feat(channels): AggregateChannel unifies BinOp + DeltaChannel with snapshot_frequency knob

New `AggregateChannel(operator, *, snapshot_frequency=1, typ=None)` replaces
the experimental `DeltaChannel`. `snapshot_frequency=1` (default) is today's
`BinaryOperatorAggregate`. `snapshot_frequency=math.inf` is today's pure-delta
behavior. Integer values between bound replay depth — deep-thread reads
become O(snapshot_frequency) instead of O(thread depth).

- `BinaryOperatorAggregate` is now a thin subclass (snapshot_frequency=1),
  preserving isinstance checks and `_is_field_binop` detection.
- `DeltaChannel` (private, experimental, underscored module) is removed.
  Migration: `AggregateChannel(op, snapshot_frequency=math.inf)`.
- `create_checkpoint` is step-aware: non-snapshot steps store DELTA_SENTINEL;
  snapshot steps store the full blob.
- `_get_channel_writes_history` walk is fixed: pending_writes of a
  terminator ancestor encode its state→child transition and are now
  collected BEFORE checking the blob (old code silently dropped them,
  which was hidden because no prior scenario had a FULL blob mid-thread).

Saver API is unchanged. Batched multi-channel walks, `walk_writes`/
`put_channel_snapshot` refactor, `coalesce=` kwarg, and Option A
(channel_versions delta-encoding) are deferred per spec. Design:
`docs/superpowers/specs/2026-04-24-aggregate-channel-design.md`.

Verified:
- snapshot_frequency ∈ {1, 2, 3, 5, 10, math.inf} all reconstruct correctly
  on a 7-invoke / 14-message thread.
- Multi-channel graph with different snapshot_frequency per channel works.
- All 10 migration tests pass unchanged (pre-BinOp-to-Delta migration path).
- Channel unit tests pass (35 tests).

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

docs/superpowers/specs/2026-04-24-aggregate-channel-design.md

@@ -0,0 +1,219 @@
+# AggregateChannel — unified fold-reducer channel with configurable snapshot cadence
+
+**Status:** MVP scope approved. Implementation starting 2026-04-24.
+**Supersedes:** `langgraph.channels._delta.DeltaChannel` (experimental, private).
+**Branch:** `sr/even-better-writes-idea` (forked from `delta-channel-writes-based`).
+
+## Problem
+
+The experimental `DeltaChannel` (introduced earlier on this branch) stores a
+sentinel in every checkpoint and reconstructs state by walking ancestor
+writes. It eliminates O(N²) blob growth for append-style reducers on long
+threads, but read cost now scales O(N) with thread depth — every load
+replays every write since the start of the thread.
+
+`BinaryOperatorAggregate` is the opposite extreme: always snapshots the full
+value every step. Zero replay cost at read time, but O(N²) storage on
+append-heavy workloads.
+
+These are endpoints of the same axis. A single channel class parameterised
+on snapshot cadence covers both — plus every intermediate point.
+
+The concrete pain that surfaced this design: deep-agent workloads at
+200+ turns pay O(200) replay per read under `DeltaChannel`. A
+`snapshot_frequency` knob bounds that to O(snapshot_frequency) regardless
+of thread depth.
+
+## MVP scope (this PR)
+
+1. **New class `AggregateChannel`** at `libs/langgraph/langgraph/channels/aggregate.py`:
+
+   ```python
+   class AggregateChannel(Generic[Value], BaseChannel[Any, Any, Any]):
+       def __init__(
+           self,
+           operator: Callable[[Value, Value], Value],
+           *,
+           snapshot_frequency: int | float = 1,
+           typ: type[Value] | None = None,
+       ): ...
+   ```
+
+   - `snapshot_frequency=1` (default): full snapshot every step. Equivalent
+     to today's `BinaryOperatorAggregate`.
+   - `snapshot_frequency=N` (integer > 1): full snapshot every Nth step;
+     sentinel on other steps.
+   - `snapshot_frequency=math.inf`: never snapshot. Equivalent to today's
+     `DeltaChannel`.
+   - `typ` inferred from `Annotated[...]` via `_strip_extras` when used in
+     a `TypedDict` state schema; kwarg is the escape hatch for imperative
+     constructions.
+
+2. **Rewire `BinaryOperatorAggregate` as a subclass** of `AggregateChannel`
+   with `snapshot_frequency=1` hard-coded. Preserves
+   `isinstance(x, BinaryOperatorAggregate)` for any existing user code and
+   `_is_field_binop` detection in `graph/state.py`.
+
+   ```python
+   class BinaryOperatorAggregate(AggregateChannel):
+       def __init__(self, typ, operator):
+           super().__init__(operator, typ=typ, snapshot_frequency=1)
+   ```
+
+3. **Delete `langgraph/channels/_delta.py`** (`DeltaChannel`). It was
+   experimental, private, underscored, and not re-exported — clean
+   removal. Users migrate to `AggregateChannel(op, snapshot_frequency=math.inf)`.
+
+4. **Step-aware `create_checkpoint`** at `libs/langgraph/langgraph/pregel/_checkpoint.py`:
+
+   `AggregateChannel` exposes a helper method:
+
+   ```python
+   def is_snapshot_step(self, step: int) -> bool:
+       if self.snapshot_frequency == 1:
+           return True
+       if self.snapshot_frequency == math.inf:
+           return False
+       return step % self.snapshot_frequency == 0
+   ```
+
+   `create_checkpoint` calls this per channel. When it returns `False`,
+   the row stores `DELTA_SENTINEL` for that channel; when `True`, it
+   stores `ch.checkpoint()` as today. `step` is already an argument to
+   `create_checkpoint` — no new threading. The explicit branches on
+   `1` and `math.inf` avoid relying on `step % math.inf` NaN arithmetic
+   and make the common cases (always snapshot / never snapshot) free of
+   a modulo.
+
+5. **Generalise `channels_from_checkpoint`**. The existing DeltaChannel-specific
+   branch keys off `isinstance(spec, DeltaChannel)`; change to
+   `isinstance(spec, AggregateChannel) and spec.snapshot_frequency != 1`.
+   The existing "pre-delta seed terminator" walk already treats any
+   non-sentinel ancestor blob as the base value and stops — no change
+   needed to saver-side logic. A `snapshot_frequency=10` blob at step 50
+   serves as a natural terminator for a walk starting at step 57.
+
+6. **Saver API stays as-is.** `_get_channel_writes_history` (private,
+   underscored) continues to be the reconstruction hook. The broader
+   refactor (`walk_writes` + `put_channel_snapshot`, batched multi-channel
+   walks) is deferred to a follow-up PR that can benchmark its own win
+   independently.
+
+## Explicitly deferred (documented, not implemented here)
+
+Each of the below lands as its own PR on top of this MVP.
+
+- **`coalesce=` kwarg on `AggregateChannel`.** Batch-shape reducer for users
+  who need to see all of a step's writes at once (non-binary-foldable
+  reducers: median, priority-pick, dedup-across-writes). Additive to the
+  existing `operator` kwarg; exactly one of the two must be provided.
+- **Saver API boundary refactor.** Rename `_get_channel_writes_history` →
+  `walk_writes(config, *, channels=None)`. Move the `DELTA_SENTINEL`
+  terminator check from saver-side to pregel-side. Saver becomes a pure
+  storage primitive (aligns with every event-sourced system surveyed:
+  Akka Persistence, EventStoreDB, Kafka Streams, Postgres logical
+  replication, Firestore). Research memo captured in brainstorming session.
+- **Batched multi-channel walks.** One ancestor-walk query per read
+  regardless of how many `AggregateChannel` channels need hydration.
+  Today each channel triggers its own walk; deep-agent graphs with 7 state
+  channels pay 7× the latency.
+- **`put_channel_snapshot` saver hook.** Opportunistic/manual compaction of
+  a pure-delta (`snapshot_frequency=math.inf`) thread by retroactively
+  promoting a sentinel row to a full blob. Separate from the write hot path.
+- **ShallowPostgresSaver compat.** `snapshot_frequency > 1` is fundamentally
+  incompatible with shallow savers (no parent chain → nowhere to walk).
+  Detect at attach time and error loudly. The existing `DeltaChannel` has
+  the same silent incompatibility today; make it explicit in the same
+  pass.
+- **Option A — channel_versions / versions_seen delta-encoding.** Documented
+  in `notes/delta_checkpoint_rows.md`. 60% win on the checkpoint row table,
+  reuses the same parent-walk machinery. Requires `Checkpoint.v` bump
+  (4 → 5), so wants to land on top of the saver API refactor, not stacked
+  with this MVP.
+
+## Key design decisions and why
+
+- **`operator`-only MVP, `coalesce` deferred.** The deepagents workload
+  uses `add_messages`-shape reducers (binary-foldable). Shipping
+  `coalesce=` now expands the API surface before we've validated that
+  the snapshot-cadence half works on a real workload. `coalesce=` is
+  additive and can land without breaking anyone.
+- **Subclass, not alias.** `BinaryOperatorAggregate` is imported and
+  instantiated directly in at least `libs/langgraph/langgraph/graph/state.py:1711`
+  (`_is_field_binop`). A factory function breaks `isinstance`; a subclass
+  doesn't.
+- **`snapshot_frequency`, not `snapshot_every`.** Chosen by user preference;
+  semantically identical (integer period, default 1).
+- **No saver API change in MVP.** The saver's existing
+  `_get_channel_writes_history` contract is sufficient for the cadence
+  knob to work. Deferring the saver refactor lets this PR land
+  independently and the refactor benchmark against a stable baseline.
+- **No benchmark harness in this PR.** Benchmarking happens externally
+  against deepagents.
+- **DELTA_SENTINEL keeps its name.** Even though the class renames to
+  `AggregateChannel`, the sentinel itself is still "this row represents a
+  delta from ancestors" — the name is accurate. Rename could happen in a
+  later cleanup if desired but isn't in scope.
+
+## Migration semantics
+
+- **Existing threads with `BinaryOperatorAggregate`** continue to work
+  unchanged — they're now instances of `AggregateChannel` with
+  `snapshot_frequency=1`, and the runtime code paths are identical.
+- **Switching `snapshot_frequency` mid-thread** (e.g. user bumps
+  `snapshot_frequency=1` → `10` on an existing thread): pre-change
+  checkpoints have full blobs; they act as natural walk terminators for
+  post-change reads. No explicit migration step. No data loss.
+- **Reverse migration** (`snapshot_frequency=10` → `1`): next write produces
+  a full blob. Reads at ancestors still find the right terminator. Safe.
+- **Switching `snapshot_frequency=math.inf` → any finite value:**
+  next snapshot-step writes a full blob that closes all prior sentinel-only
+  ancestry. Walks from that point forward stop at the new base rather
+  than walking to the root.
+- **External users who imported `langgraph.channels._delta.DeltaChannel`**:
+  `ImportError` at upgrade time. Underscored + experimental + docstring
+  says "subject to change or removal without notice" — documented
+  breakage; migration is a one-line swap.
+
+## File-level change list
+
+**New:**
+- `libs/langgraph/langgraph/channels/aggregate.py` — `AggregateChannel` class
+
+**Modified:**
+- `libs/langgraph/langgraph/channels/binop.py` — `BinaryOperatorAggregate`
+  becomes subclass of `AggregateChannel`; reducer logic moves to base.
+- `libs/langgraph/langgraph/channels/__init__.py` — export
+  `AggregateChannel`.
+- `libs/langgraph/langgraph/pregel/_checkpoint.py`:
+  - `create_checkpoint`: step-aware sentinel vs blob decision.
+  - `channels_from_checkpoint` / `achannels_from_checkpoint`: key off
+    `AggregateChannel` instead of `DeltaChannel`.
+  - `DeltaChannel` import removed.
+- `libs/langgraph/langgraph/graph/state.py` — `_is_field_binop` continues
+  to work unchanged (subclass relationship preserves detection).
+
+**Deleted:**
+- `libs/langgraph/langgraph/channels/_delta.py`
+
+**Tests updated:**
+- Any test importing `DeltaChannel` from `langgraph.channels._delta` —
+  switch to `AggregateChannel(op, snapshot_frequency=math.inf)`.
+- Add: parity test for `snapshot_frequency=1` vs today's `BinaryOperatorAggregate`
+  on the same workload.
+- Add: cadence test at `snapshot_frequency=10` on a 50-step thread —
+  verify blobs land on steps 0/10/20/30/40/50, sentinels elsewhere, and
+  reads at every step produce the same value as the all-snapshot baseline.
+- Add: mid-thread `snapshot_frequency` change — verify no data loss.
+
+## Out of scope
+
+- Performance benchmarking (user runs externally on deepagents).
+- Documentation/tutorial updates for the new knob (until MVP validates).
+- Any saver-side changes.
+- Any `Checkpoint.v` bump.
+- Public API promotion — `AggregateChannel` replaces a private
+  experimental class; it's immediately public by virtue of living in
+  `langgraph.channels`, but the `snapshot_frequency > 1` path inherits
+  DeltaChannel's "experimental, validate on real workloads first"
+  caveat until benchmark confirms it.

libs/checkpoint-postgres/langgraph/checkpoint/postgres/__init__.py

@@ -4,7 +4,7 @@ import threading
 from collections import defaultdict
 from collections.abc import Iterator, Sequence
 from contextlib import contextmanager
-from typing import Any, cast
+from typing import Any
 
 from langchain_core.runnables import RunnableConfig
 from langgraph.checkpoint.base import (
@@ -26,9 +26,10 @@ from psycopg_pool import ConnectionPool
 
 from langgraph.checkpoint.postgres import _internal
 from langgraph.checkpoint.postgres.base import (
-    SELECT_DELTA_COMBINED_SQL,
+    SELECT_DELTA_BLOBS_SQL,
+    SELECT_DELTA_PARENTS_SQL,
+    SELECT_DELTA_WRITES_SQL,
     BasePostgresSaver,
-    _DeltaCombinedRow,
 )
 from langgraph.checkpoint.postgres.shallow import ShallowPostgresSaver
 
@@ -441,9 +442,10 @@ class PostgresSaver(BasePostgresSaver):
     ) -> _ChannelWritesHistory:
         """Fast-path override of `BaseCheckpointSaver._get_channel_writes_history`.
 
-        One combined UNION ALL query (`SELECT_DELTA_COMBINED_SQL`) fetches rows
-        from `checkpoints`, `checkpoint_writes`, and `checkpoint_blobs` in a
-        single roundtrip; the ancestor walk runs in Python.
+        Three indexed roundtrips (`checkpoints`, `checkpoint_writes`,
+        `checkpoint_blobs`) each filtered by `(thread_id, checkpoint_ns)` and
+        the per-table key. Plain SELECTs let the planner pick straight index
+        scans; rationale + benchmark in `notes/delta_channel_query_bench.md`.
         """
         thread_id = config["configurable"]["thread_id"]
         checkpoint_ns = config["configurable"].get("checkpoint_ns", "")
@@ -457,25 +459,18 @@ class PostgresSaver(BasePostgresSaver):
                 return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
             checkpoint_id = target.config["configurable"]["checkpoint_id"]
         with self._cursor() as cur:
-            cur.execute(
-                SELECT_DELTA_COMBINED_SQL,
-                (
-                    channel,
-                    thread_id,
-                    checkpoint_ns,
-                    thread_id,
-                    checkpoint_ns,
-                    channel,
-                    thread_id,
-                    checkpoint_ns,
-                    channel,
-                ),
-            )
-            rows = cur.fetchall()
+            cur.execute(SELECT_DELTA_PARENTS_SQL, (channel, thread_id, checkpoint_ns))
+            parents_rows = cur.fetchall()
+            cur.execute(SELECT_DELTA_WRITES_SQL, (thread_id, checkpoint_ns, channel))
+            writes_rows = cur.fetchall()
+            cur.execute(SELECT_DELTA_BLOBS_SQL, (thread_id, checkpoint_ns, channel))
+            blobs_rows = cur.fetchall()
         return self._build_delta_channel_writes_history(
             channel=channel,
             target_id=checkpoint_id,
-            rows=cast("list[_DeltaCombinedRow]", rows),
+            parents_rows=parents_rows,
+            writes_rows=writes_rows,
+            blobs_rows=blobs_rows,
         )
 
     def _load_checkpoint_tuple(self, value: DictRow) -> CheckpointTuple:
@@ -490,6 +485,7 @@ class PostgresSaver(BasePostgresSaver):
             including its configuration, metadata, parent checkpoint (if any),
             and pending writes.
         """
+        channel_values = self._load_blobs(value["channel_values"])
         return CheckpointTuple(
             {
                 "configurable": {
@@ -502,7 +498,7 @@ class PostgresSaver(BasePostgresSaver):
                 **value["checkpoint"],
                 "channel_values": {
                     **(value["checkpoint"].get("channel_values") or {}),
-                    **self._load_blobs(value["channel_values"]),
+                    **channel_values,
                 },
             },
             value["metadata"],

libs/checkpoint-postgres/langgraph/checkpoint/postgres/aio.py

@@ -4,7 +4,7 @@ import asyncio
 from collections import defaultdict
 from collections.abc import AsyncIterator, Iterator, Sequence
 from contextlib import asynccontextmanager
-from typing import Any, cast
+from typing import Any
 
 from langchain_core.runnables import RunnableConfig
 from langgraph.checkpoint.base import (
@@ -26,9 +26,10 @@ from psycopg_pool import AsyncConnectionPool
 
 from langgraph.checkpoint.postgres import _ainternal
 from langgraph.checkpoint.postgres.base import (
-    SELECT_DELTA_COMBINED_SQL,
+    SELECT_DELTA_BLOBS_SQL,
+    SELECT_DELTA_PARENTS_SQL,
+    SELECT_DELTA_WRITES_SQL,
     BasePostgresSaver,
-    _DeltaCombinedRow,
 )
 from langgraph.checkpoint.postgres.shallow import AsyncShallowPostgresSaver
 
@@ -402,10 +403,10 @@ class AsyncPostgresSaver(BasePostgresSaver):
     ) -> _ChannelWritesHistory:
         """Fast-path override of `BaseCheckpointSaver._aget_channel_writes_history`.
 
-        One combined UNION ALL query (`SELECT_DELTA_COMBINED_SQL`) fetches rows
-        from `checkpoints`, `checkpoint_writes`, and `checkpoint_blobs` in a
-        single roundtrip; rows are assembled by the shared pure helper on
-        `BasePostgresSaver`.
+        Three indexed roundtrips (`checkpoints`, `checkpoint_writes`,
+        `checkpoint_blobs`); rows assembled by the shared pure helper on
+        `BasePostgresSaver`. Rationale + benchmark in
+        `notes/delta_channel_query_bench.md`.
         """
         thread_id = config["configurable"]["thread_id"]
         checkpoint_ns = config["configurable"].get("checkpoint_ns", "")
@@ -417,24 +418,23 @@ class AsyncPostgresSaver(BasePostgresSaver):
             checkpoint_id = target.config["configurable"]["checkpoint_id"]
         async with self._cursor() as cur:
             await cur.execute(
-                SELECT_DELTA_COMBINED_SQL,
-                (
-                    channel,
-                    thread_id,
-                    checkpoint_ns,
-                    thread_id,
-                    checkpoint_ns,
-                    channel,
-                    thread_id,
-                    checkpoint_ns,
-                    channel,
-                ),
+                SELECT_DELTA_PARENTS_SQL, (channel, thread_id, checkpoint_ns)
             )
-            rows = await cur.fetchall()
+            parents_rows = await cur.fetchall()
+            await cur.execute(
+                SELECT_DELTA_WRITES_SQL, (thread_id, checkpoint_ns, channel)
+            )
+            writes_rows = await cur.fetchall()
+            await cur.execute(
+                SELECT_DELTA_BLOBS_SQL, (thread_id, checkpoint_ns, channel)
+            )
+            blobs_rows = await cur.fetchall()
         return self._build_delta_channel_writes_history(
             channel=channel,
             target_id=checkpoint_id,
-            rows=cast("list[_DeltaCombinedRow]", rows),
+            parents_rows=parents_rows,
+            writes_rows=writes_rows,
+            blobs_rows=blobs_rows,
         )
 
     async def _load_checkpoint_tuple(self, value: DictRow) -> CheckpointTuple:
@@ -449,11 +449,18 @@ class AsyncPostgresSaver(BasePostgresSaver):
             including its configuration, metadata, parent checkpoint (if any),
             and pending writes.
         """
+        thread_id = value["thread_id"]
+        checkpoint_ns = value["checkpoint_ns"]
+        blob_values = value["channel_values"]
+        channel_values: dict[str, Any] = {}
+        if blob_values:
+            channel_values = self._load_blobs(blob_values)
+
         return CheckpointTuple(
             {
                 "configurable": {
-                    "thread_id": value["thread_id"],
-                    "checkpoint_ns": value["checkpoint_ns"],
+                    "thread_id": thread_id,
+                    "checkpoint_ns": checkpoint_ns,
                     "checkpoint_id": value["checkpoint_id"],
                 }
             },
@@ -461,15 +468,15 @@ class AsyncPostgresSaver(BasePostgresSaver):
                 **value["checkpoint"],
                 "channel_values": {
                     **(value["checkpoint"].get("channel_values") or {}),
-                    **self._load_blobs(value["channel_values"]),
+                    **channel_values,
                 },
             },
             value["metadata"],
             (
                 {
                     "configurable": {
-                        "thread_id": value["thread_id"],
-                        "checkpoint_ns": value["checkpoint_ns"],
+                        "thread_id": thread_id,
+                        "checkpoint_ns": checkpoint_ns,
                         "checkpoint_id": value["parent_checkpoint_id"],
                     }
                 }

libs/checkpoint-postgres/langgraph/checkpoint/postgres/base.py

@@ -4,7 +4,7 @@ import random
 import warnings
 from collections.abc import Sequence
 from importlib.metadata import version as get_version
-from typing import Any, TypedDict, cast
+from typing import Any, cast
 
 from langchain_core.runnables import RunnableConfig
 from langgraph.checkpoint.base import (
@@ -155,58 +155,26 @@ INSERT_CHECKPOINT_WRITES_SQL = """
     ON CONFLICT (thread_id, checkpoint_ns, checkpoint_id, task_id, idx) DO NOTHING
 """
 
-
-class _DeltaCombinedRow(TypedDict, total=False):
-    """One row from `SELECT_DELTA_COMBINED_SQL` (a UNION ALL of three tables).
-
-    Every row carries `_kind` ("p" / "w" / "b") plus whichever columns are
-    relevant for that kind; irrelevant columns are NULL and typed as `None`.
-    """
-
-    _kind: str  # always present: "p", "w", or "b"
-    # checkpoint row ("p")
-    checkpoint_id: str | None
-    parent_checkpoint_id: str | None
-    ver: str | None
-    # write / blob rows ("w", "b")
-    type: str | None
-    blob: bytes | None
-    # write row only ("w")
-    task_id: str | None
-    idx: int | None
-    # blob row only ("b")
-    version: str | None
-
-
-# DeltaChannel reconstruction: one UNION ALL query fetches checkpoints,
-# writes, and blobs for `channel` in one roundtrip; the ancestor walk runs
-# in Python in `_build_delta_channel_writes_history`.
-#
-# Parameter order: (channel, thread_id, checkpoint_ns,
-#                   thread_id, checkpoint_ns, channel,
-#                   thread_id, checkpoint_ns, channel)
-SELECT_DELTA_COMBINED_SQL = """
-    SELECT 'p'::text       AS _kind,
-           checkpoint_id,
+# DeltaChannel reconstruction: three plain indexed SELECTs per channel.
+# Bench (notes/delta_channel_query_bench.md) showed the prior recursive CTE
+# carried a hidden O(ancestors x blobs_in_thread) join; plain SELECTs are
+# 3x-100x faster in the realistic depth range and the Python walk is O(n).
+SELECT_DELTA_PARENTS_SQL = """
+    SELECT checkpoint_id,
            parent_checkpoint_id,
-           checkpoint -> 'channel_versions' ->> %s AS ver,
-           NULL::text       AS type,
-           NULL::bytea      AS blob,
-           NULL::text       AS task_id,
-           NULL::int        AS idx,
-           NULL::text       AS version
+           checkpoint -> 'channel_versions' ->> %s AS ver
     FROM checkpoints
     WHERE thread_id = %s AND checkpoint_ns = %s
-    UNION ALL
-    SELECT 'w',
-           checkpoint_id, NULL, NULL,
-           type, blob, task_id, idx, NULL
+"""
+
+SELECT_DELTA_WRITES_SQL = """
+    SELECT checkpoint_id, type, blob, task_id, idx
     FROM checkpoint_writes
     WHERE thread_id = %s AND checkpoint_ns = %s AND channel = %s
-    UNION ALL
-    SELECT 'b',
-           NULL, NULL, NULL,
-           type, blob, NULL, NULL, version
+"""
+
+SELECT_DELTA_BLOBS_SQL = """
+    SELECT version, type, blob
     FROM checkpoint_blobs
     WHERE thread_id = %s AND checkpoint_ns = %s AND channel = %s
 """
@@ -244,28 +212,32 @@ class BasePostgresSaver(BaseCheckpointSaver[str]):
         )
 
     def _load_blobs(
-        self, blob_values: list[tuple[bytes, bytes, bytes]]
+        self,
+        blob_values: Any,
     ) -> dict[str, Any]:
         if not blob_values:
             return {}
-        return {
-            k.decode(): self.serde.loads_typed((t.decode(), v))
-            for k, t, v in blob_values
-            if t.decode() != "empty"
-        }
+        result: dict[str, Any] = {}
+        for k, t, v in blob_values:
+            type_tag = t.decode()
+            if type_tag != "empty":
+                result[k.decode()] = self.serde.loads_typed((type_tag, v))
+        return result
 
     def _build_delta_channel_writes_history(
         self,
         *,
         channel: str,
         target_id: str,
-        rows: Sequence[_DeltaCombinedRow],
+        parents_rows: Sequence[Any],
+        writes_rows: Sequence[Any],
+        blobs_rows: Sequence[Any],
     ) -> _ChannelWritesHistory:
-        """Reconstruct one delta channel's history from the combined UNION ALL rows.
+        """Reconstruct one delta channel's history from rows of the three SELECTs.
 
         Pure data transform shared by sync (`PostgresSaver`) and async
-        (`AsyncPostgresSaver`); both paths run `SELECT_DELTA_COMBINED_SQL`
-        and feed the tagged rows here.
+        (`AsyncPostgresSaver`); both paths run the queries themselves and
+        feed the rows here.
 
         Walk is newest → oldest from the target's parent. A non-sentinel
         blob in `checkpoint_blobs` (a pre-delta snapshot) terminates the
@@ -276,49 +248,47 @@ class BasePostgresSaver(BaseCheckpointSaver[str]):
         """
         parent_of: dict[str, str | None] = {}
         ver_of: dict[str, str | None] = {}
+        for r in parents_rows:
+            cid = r["checkpoint_id"]
+            parent_of[cid] = r["parent_checkpoint_id"]
+            ver_of[cid] = r["ver"]
+
+        ancestors: list[str] = []
+        cid = parent_of.get(target_id)
+        while cid is not None:
+            ancestors.append(cid)
+            cid = parent_of.get(cid)
+        if not ancestors:
+            return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
+        ancestor_set = set(ancestors)
+
+        # Group writes by ancestor cid; sort within (task_id DESC, idx DESC)
+        # to match the prior CTE ordering — newest write first per ancestor.
         writes_by_cid: dict[str, list[tuple[str, bytes, str, int]]] = {}
-        blob_by_ver: dict[str, tuple[str, bytes]] = {}
-
-        for r in rows:
-            kind = r["_kind"]
-            if kind == "p":
-                cid = cast(str, r["checkpoint_id"])
-                parent_of[cid] = r["parent_checkpoint_id"]
-                ver_of[cid] = r["ver"]
-            elif kind == "w":
-                cid = cast(str, r["checkpoint_id"])
-                writes_by_cid.setdefault(cid, []).append(
-                    cast(
-                        "tuple[str, bytes, str, int]",
-                        (r["type"], r["blob"], r["task_id"], r["idx"]),
-                    )
-                )
-            else:  # kind == "b"
-                blob_by_ver[cast(str, r["version"])] = cast(
-                    "tuple[str, bytes]", (r["type"], r["blob"])
-                )
-
-        # newest write first per ancestor (task_id DESC, idx DESC)
+        for r in writes_rows:
+            cid = r["checkpoint_id"]
+            if cid not in ancestor_set:
+                continue
+            writes_by_cid.setdefault(cid, []).append(
+                (r["type"], r["blob"], r["task_id"], r["idx"])
+            )
         for ws in writes_by_cid.values():
             ws.sort(key=lambda w: (w[2], w[3]), reverse=True)
 
-        ancestors: list[str] = []
-        cur_cid: str | None = parent_of.get(target_id)
-        while cur_cid is not None:
-            ancestors.append(cur_cid)
-            cur_cid = parent_of.get(cur_cid)
-        if not ancestors:
-            return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
+        blob_by_ver: dict[str, tuple[str, bytes]] = {
+            r["version"]: (r["type"], r["blob"]) for r in blobs_rows
+        }
 
         collected: list[PendingWrite] = []  # newest first; reversed at the end
         for cid in ancestors:
-            # Collect writes first — they encode the transition FROM this
-            # ancestor's state to its child's and must be included even if
-            # this ancestor is also the seed checkpoint.
+            # Collect this ancestor's pending_writes FIRST. They encode the
+            # transition from state-AT-this-ancestor to state-AT-its-child;
+            # the ancestor's blob only reflects state AT the ancestor, not
+            # post-transition. Both pre-delta migration and snapshot-cadence
+            # cases require these writes folded onto the seed.
             for type_tag, write_blob, task_id, _idx in writes_by_cid.get(cid, []):
                 val = self.serde.loads_typed((type_tag, write_blob))
                 collected.append((task_id, channel, val))
-            # Then check seed terminator.
             ver = ver_of.get(cid)
             if ver is not None:
                 seed_blob = blob_by_ver.get(ver)

libs/checkpoint-postgres/pyproject.toml

@@ -12,7 +12,7 @@ readme = "README.md"
 license = "MIT"
 license-files = ['LICENSE']
 dependencies = [
-  "langgraph-checkpoint>=4.0.3,<5.0.0",
+  "langgraph-checkpoint>=2.1.2,<5.0.0",
   "orjson>=3.11.5",
   "psycopg>=3.2.0",
   "psycopg-pool>=3.2.0",

libs/checkpoint-postgres/tests/test_async.py

@@ -377,19 +377,19 @@ async def test_get_checkpoint_no_channel_values(
 async def test_delta_channel_chain_reconstruction(saver_name: str) -> None:
     """AsyncPostgresSaver reconstructs DeltaChannel chain via point-lookup traversal."""
     pytest.importorskip(
-        "langgraph.channels.delta", reason="langgraph core not installed"
+        "langgraph.channels._delta", reason="langgraph core not installed"
     )
 
     from typing import Annotated
 
     from langchain_core.messages import AIMessage, HumanMessage
-    from langgraph.channels.delta import DeltaChannel
+    from langgraph.channels._delta import DeltaChannel
     from langgraph.graph import START, StateGraph
-    from langgraph.graph.message import _messages_delta_reducer
+    from langgraph.graph.message import add_messages
     from typing_extensions import TypedDict
 
     class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
+        messages: Annotated[list, DeltaChannel(add_messages)]
 
     def respond(state: State) -> dict:
         n = len(state["messages"])

libs/checkpoint/langgraph/checkpoint/base/__init__.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import contextvars
 import copy
 import logging
 from collections.abc import AsyncIterator, Collection, Iterator, Mapping, Sequence
@@ -32,6 +33,14 @@ from langgraph.checkpoint.serde.types import (
 V = TypeVar("V", int, float, str)
 PendingWrite = tuple[str, str, Any]
 
+# Task-local guard: ContextVar is copied per asyncio Task, so concurrent
+# requests on the same event-loop thread do not share this flag. A plain
+# `threading.local()` would leak across tasks and let one in-flight
+# reconstruction silently short-circuit another.
+_DELTA_RECONSTRUCTION: contextvars.ContextVar[bool] = contextvars.ContextVar(
+    "_DELTA_RECONSTRUCTION", default=False
+)
+
 
 logger = logging.getLogger(__name__)
 
@@ -486,20 +495,6 @@ class BaseCheckpointSaver(Generic[V]):
         """
         raise NotImplementedError
 
-    def _get_tuple_raw(self, config: RunnableConfig) -> CheckpointTuple | None:
-        """Pure storage read used by `_get_channel_writes_history`.
-
-        Must return the same value as `get_tuple` but must NOT trigger channel
-        reconstruction; otherwise the channel-hydration path would re-enter
-        `_get_channel_writes_history`. Override only if `get_tuple` itself
-        performs channel hydration.
-        """
-        return self.get_tuple(config)
-
-    async def _aget_tuple_raw(self, config: RunnableConfig) -> CheckpointTuple | None:
-        """Async version of `_get_tuple_raw`. See docstring there."""
-        return await self.aget_tuple(config)
-
     def _get_channel_writes_history(
         self, config: RunnableConfig, channel: str
     ) -> _ChannelWritesHistory:
@@ -528,61 +523,85 @@ class BaseCheckpointSaver(Generic[V]):
 
         Underscore-prefixed because the method surface is experimental.
         """
-        collected: list[PendingWrite] = []  # newest first; reversed at the end
-        target_tuple = self._get_tuple_raw(config)
-        cursor_config: RunnableConfig | None = (
-            target_tuple.parent_config if target_tuple else None
-        )
-        while cursor_config is not None:
-            tup = self._get_tuple_raw(cursor_config)
-            if tup is None:
-                break
-            # Collect this ancestor's writes FIRST — they encode the
-            # transition from this ancestor's state to its child's, so
-            # they must be included whether or not this ancestor is the
-            # seed terminator.
-            if tup.pending_writes:
-                # Within a superstep, pending_writes are oldest→newest;
-                # reverse to scan newest-first.
-                for write in reversed(tup.pending_writes):
-                    if write[1] != channel:
-                        continue
-                    collected.append(write)
-            # Seed terminator: any non-sentinel blob on an ancestor
-            # establishes the reconstruction base. Stop here.
-            ancestor_value = tup.checkpoint["channel_values"].get(channel)
-            if ancestor_value is not None and ancestor_value is not DELTA_SENTINEL:
-                collected.reverse()
-                return _ChannelWritesHistory(seed=ancestor_value, writes=collected)
-            cursor_config = tup.parent_config
-        collected.reverse()
-        return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
+        # Guard against re-entrant calls: when get_tuple() triggers
+        # reconstruction which calls get_tuple() again, the inner call
+        # short-circuits here.
+        if _DELTA_RECONSTRUCTION.get():
+            return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
+
+        token = _DELTA_RECONSTRUCTION.set(True)
+        try:
+            collected: list[PendingWrite] = []  # newest first; reversed at the end
+            target_tuple = self.get_tuple(config)
+            cursor_config: RunnableConfig | None = (
+                target_tuple.parent_config if target_tuple else None
+            )
+            while cursor_config is not None:
+                tup = self.get_tuple(cursor_config)
+                if tup is None:
+                    break
+                # Collect this ancestor's pending_writes FIRST. They encode
+                # the transition from this ancestor's state to its child's
+                # state (K → K+1); the ancestor's `channel_values[channel]`
+                # blob reflects state AT K only, not post-transition. Both
+                # the pre-delta migration case and the snapshot-cadence
+                # case require these writes to be included.
+                if tup.pending_writes:
+                    # Within a superstep, pending_writes are oldest→newest;
+                    # reverse to scan newest-first.
+                    for write in reversed(tup.pending_writes):
+                        if write[1] != channel:
+                            continue
+                        collected.append(write)
+                # Seed terminator: any non-sentinel blob on an ancestor
+                # establishes the reconstruction base (state AT K). The
+                # writes we just collected fold on top to produce state
+                # at K+1; subsequent (already-collected, child-side)
+                # writes fold the chain up to the target.
+                ancestor_value = tup.checkpoint["channel_values"].get(channel)
+                if ancestor_value is not None and ancestor_value is not DELTA_SENTINEL:
+                    collected.reverse()
+                    return _ChannelWritesHistory(seed=ancestor_value, writes=collected)
+                cursor_config = tup.parent_config
+            collected.reverse()
+            return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
+        finally:
+            _DELTA_RECONSTRUCTION.reset(token)
 
     async def _aget_channel_writes_history(
         self, config: RunnableConfig, channel: str
     ) -> _ChannelWritesHistory:
         """Async version of `_get_channel_writes_history`. See docstring there."""
-        collected: list[PendingWrite] = []
-        target_tuple = await self._aget_tuple_raw(config)
-        cursor_config: RunnableConfig | None = (
-            target_tuple.parent_config if target_tuple else None
-        )
-        while cursor_config is not None:
-            tup = await self._aget_tuple_raw(cursor_config)
-            if tup is None:
-                break
-            if tup.pending_writes:
-                for write in reversed(tup.pending_writes):
-                    if write[1] != channel:
-                        continue
-                    collected.append(write)
-            ancestor_value = tup.checkpoint["channel_values"].get(channel)
-            if ancestor_value is not None and ancestor_value is not DELTA_SENTINEL:
-                collected.reverse()
-                return _ChannelWritesHistory(seed=ancestor_value, writes=collected)
-            cursor_config = tup.parent_config
-        collected.reverse()
-        return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
+        if _DELTA_RECONSTRUCTION.get():
+            return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=[])
+
+        token = _DELTA_RECONSTRUCTION.set(True)
+        try:
+            collected: list[PendingWrite] = []
+            target_tuple = await self.aget_tuple(config)
+            cursor_config: RunnableConfig | None = (
+                target_tuple.parent_config if target_tuple else None
+            )
+            while cursor_config is not None:
+                tup = await self.aget_tuple(cursor_config)
+                if tup is None:
+                    break
+                # See sync variant for rationale: collect pending_writes
+                # BEFORE checking the blob terminator.
+                if tup.pending_writes:
+                    for write in reversed(tup.pending_writes):
+                        if write[1] != channel:
+                            continue
+                        collected.append(write)
+                ancestor_value = tup.checkpoint["channel_values"].get(channel)
+                if ancestor_value is not None and ancestor_value is not DELTA_SENTINEL:
+                    collected.reverse()
+                    return _ChannelWritesHistory(seed=ancestor_value, writes=collected)
+                cursor_config = tup.parent_config
+            collected.reverse()
+            return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
+        finally:
+            _DELTA_RECONSTRUCTION.reset(token)
 
     def get_next_version(self, current: V | None, channel: None) -> V:
         """Generate the next version ID for a channel.

libs/checkpoint/langgraph/checkpoint/memory/__init__.py

@@ -9,7 +9,7 @@ from collections import defaultdict
 from collections.abc import AsyncIterator, Iterator, Sequence
 from contextlib import AbstractAsyncContextManager, AbstractContextManager, ExitStack
 from types import TracebackType
-from typing import Any
+from typing import Any, cast
 
 from langchain_core.runnables import RunnableConfig
 
@@ -27,7 +27,6 @@ from langgraph.checkpoint.base import (
     get_checkpoint_id,
     get_checkpoint_metadata,
 )
-from langgraph.checkpoint.serde.types import _DeltaSnapshot
 
 logger = logging.getLogger(__name__)
 
@@ -162,19 +161,26 @@ class InMemorySaver(
             chain.append(current)
             _, _, parent = entry
             current = parent
-        # Scan newest→oldest. A pre-delta blob on an ancestor terminates the
-        # walk and is bound as `seed`; without this, a thread migrated from
-        # pre-delta storage would replay ancestor writes all the way to the
-        # root AND miss any value that lived only in the old blob (e.g. from
-        # `update_state`).
-        #
-        # At each ancestor, check the blob BEFORE processing its pending
-        # writes: a pre-delta blob represents the state AT that ancestor,
-        # which already subsumes any writes stored under it. Processing
-        # those writes first would fold them into the reconstructed value
-        # twice (once via the blob, once via replay).
+        # Scan newest→oldest. At each ancestor, collect its pending_writes
+        # BEFORE checking for a non-sentinel blob terminator. Rationale:
+        # a blob at ancestor K represents state AT step K; that ancestor's
+        # pending_writes are the writes that transition state K → state K+1.
+        # Both the pre-delta migration case (pre-migration blob + post-delta
+        # child) and the snapshot-cadence case (FULL blob mid-thread + later
+        # sentinel checkpoints) need those writes folded onto the seed.
         collected: list[PendingWrite] = []  # newest first
         for cp_id in chain:  # newest → oldest
+            step_writes = self.writes.get((thread_id, checkpoint_ns, cp_id), {})
+            # Within a superstep, sorted by (task_id, idx) = oldest → newest;
+            # reverse for newest-first scan.
+            for (_task_id, _idx), (tid, ch, serialized, _) in sorted(
+                step_writes.items(), reverse=True
+            ):
+                if ch != channel:
+                    continue
+                val = self.serde.loads_typed(serialized)
+                collected.append((tid, ch, val))
+
             entry = ns_storage.get(cp_id)
             if entry is not None:
                 ckpt = self.serde.loads_typed(entry[0])
@@ -186,46 +192,13 @@ class InMemorySaver(
                     if blob_entry is not None and blob_entry[0] != "empty":
                         blob_value = self.serde.loads_typed(blob_entry)
                         if blob_value is not DELTA_SENTINEL:
-                            if isinstance(blob_value, _DeltaSnapshot):
-                                # Step-based snapshot: the blob is state AT this
-                                # ancestor, but the ancestor's pending_writes
-                                # encode the NEXT step's transition and are NOT
-                                # subsumed by the snapshot — collect them first.
-                                step_writes = self.writes.get(
-                                    (thread_id, checkpoint_ns, cp_id), {}
-                                )
-                                for (_task_id, _idx), (
-                                    tid,
-                                    ch,
-                                    serialized,
-                                    _,
-                                ) in sorted(step_writes.items(), reverse=True):
-                                    if ch != channel:
-                                        continue
-                                    collected.append(
-                                        (tid, ch, self.serde.loads_typed(serialized))
-                                    )
-                                collected.reverse()
-                                return _ChannelWritesHistory(
-                                    seed=blob_value, writes=collected
-                                )
-                            # Pre-delta blob: state AT this ancestor already
-                            # subsumes its pending_writes — skip them.
+                            # Non-sentinel blob terminator: state AT this
+                            # ancestor becomes the reconstruction seed.
                             collected.reverse()
                             return _ChannelWritesHistory(
                                 seed=blob_value, writes=collected
                             )
 
-            step_writes = self.writes.get((thread_id, checkpoint_ns, cp_id), {})
-            # Within a superstep, sorted by (task_id, idx) = oldest → newest;
-            # reverse for newest-first scan.
-            for (_task_id, _idx), (tid, ch, serialized, _) in sorted(
-                step_writes.items(), reverse=True
-            ):
-                if ch != channel:
-                    continue
-                val = self.serde.loads_typed(serialized)
-                collected.append((tid, ch, val))
         collected.reverse()
         return _ChannelWritesHistory(seed=DELTA_SENTINEL, writes=collected)
 
@@ -255,13 +228,16 @@ class InMemorySaver(
                 checkpoint, metadata, parent_checkpoint_id = saved
                 writes = self.writes[(thread_id, checkpoint_ns, checkpoint_id)].values()
                 checkpoint_: Checkpoint = self.serde.loads_typed(checkpoint)
+                channel_values = self._load_blobs(
+                    thread_id,
+                    checkpoint_ns,
+                    checkpoint_["channel_versions"],
+                )
                 return CheckpointTuple(
                     config=config,
                     checkpoint={
                         **checkpoint_,
-                        "channel_values": self._load_blobs(
-                            thread_id, checkpoint_ns, checkpoint_["channel_versions"]
-                        ),
+                        "channel_values": channel_values,
                     },
                     metadata=self.serde.loads_typed(metadata),
                     pending_writes=[
@@ -285,19 +261,26 @@ class InMemorySaver(
                 checkpoint, metadata, parent_checkpoint_id = checkpoints[checkpoint_id]
                 writes = self.writes[(thread_id, checkpoint_ns, checkpoint_id)].values()
                 checkpoint_ = self.serde.loads_typed(checkpoint)
-                return CheckpointTuple(
-                    config={
+                resolved_config = cast(
+                    RunnableConfig,
+                    {
                         "configurable": {
                             "thread_id": thread_id,
                             "checkpoint_ns": checkpoint_ns,
                             "checkpoint_id": checkpoint_id,
                         }
                     },
+                )
+                channel_values = self._load_blobs(
+                    thread_id,
+                    checkpoint_ns,
+                    checkpoint_["channel_versions"],
+                )
+                return CheckpointTuple(
+                    config=resolved_config,
                     checkpoint={
                         **checkpoint_,
-                        "channel_values": self._load_blobs(
-                            thread_id, checkpoint_ns, checkpoint_["channel_versions"]
-                        ),
+                        "channel_values": channel_values,
                     },
                     metadata=self.serde.loads_typed(metadata),
                     pending_writes=[
@@ -392,21 +375,27 @@ class InMemorySaver(
 
                     checkpoint_: Checkpoint = self.serde.loads_typed(checkpoint)
 
-                    yield CheckpointTuple(
-                        config={
+                    list_config = cast(
+                        RunnableConfig,
+                        {
                             "configurable": {
                                 "thread_id": thread_id,
                                 "checkpoint_ns": checkpoint_ns,
                                 "checkpoint_id": checkpoint_id,
                             }
                         },
+                    )
+                    channel_values = self._load_blobs(
+                        thread_id,
+                        checkpoint_ns,
+                        checkpoint_["channel_versions"],
+                    )
+
+                    yield CheckpointTuple(
+                        config=list_config,
                         checkpoint={
                             **checkpoint_,
-                            "channel_values": self._load_blobs(
-                                thread_id,
-                                checkpoint_ns,
-                                checkpoint_["channel_versions"],
-                            ),
+                            "channel_values": channel_values,
                         },
                         metadata=metadata,
                         parent_config=(

libs/checkpoint/langgraph/checkpoint/serde/jsonplus.py

@@ -33,12 +33,7 @@ from langchain_core.load.load import Reviver
 from langgraph.checkpoint.serde import _msgpack as _lg_msgpack
 from langgraph.checkpoint.serde.base import SerializerProtocol
 from langgraph.checkpoint.serde.event_hooks import emit_serde_event
-from langgraph.checkpoint.serde.types import (
-    DELTA_SENTINEL,
-    SendProtocol,
-    _DeltaSentinel,
-    _DeltaSnapshot,
-)
+from langgraph.checkpoint.serde.types import DELTA_SENTINEL, SendProtocol
 from langgraph.store.base import Item
 
 if TYPE_CHECKING:
@@ -256,6 +251,8 @@ class JsonPlusSerializer(SerializerProtocol):
     def dumps_typed(self, obj: Any) -> tuple[str, bytes]:
         if obj is None:
             return "null", EMPTY_BYTES
+        elif obj is DELTA_SENTINEL:
+            return "delta", EMPTY_BYTES
         elif isinstance(obj, bytes):
             return "bytes", obj
         elif isinstance(obj, bytearray):
@@ -282,6 +279,8 @@ class JsonPlusSerializer(SerializerProtocol):
             return ormsgpack.unpackb(
                 data_, ext_hook=self._unpack_ext_hook, option=ormsgpack.OPT_NON_STR_KEYS
             )
+        elif type_ == "delta":
+            return DELTA_SENTINEL
         elif self.pickle_fallback and type_ == "pickle":
             return pickle.loads(data_)
         else:
@@ -297,16 +296,10 @@ EXT_METHOD_SINGLE_ARG = 3
 EXT_PYDANTIC_V1 = 4
 EXT_PYDANTIC_V2 = 5
 EXT_NUMPY_ARRAY = 6
-EXT_DELTA_SNAPSHOT = 7
-EXT_DELTA_SENTINEL = 8
 
 
 def _msgpack_default(obj: Any) -> str | ormsgpack.Ext:
-    if isinstance(obj, _DeltaSnapshot):
-        return ormsgpack.Ext(EXT_DELTA_SNAPSHOT, _msgpack_enc(obj.value))
-    elif isinstance(obj, _DeltaSentinel):
-        return ormsgpack.Ext(EXT_DELTA_SENTINEL, b"")
-    elif hasattr(obj, "model_dump") and callable(obj.model_dump):  # pydantic v2
+    if hasattr(obj, "model_dump") and callable(obj.model_dump):  # pydantic v2
         return ormsgpack.Ext(
             EXT_PYDANTIC_V2,
             _msgpack_enc(
@@ -620,15 +613,7 @@ def _create_msgpack_ext_hook(
         return False
 
     def ext_hook(code: int, data: bytes) -> Any:
-        if code == EXT_DELTA_SENTINEL:
-            return DELTA_SENTINEL
-        elif code == EXT_DELTA_SNAPSHOT:
-            return _DeltaSnapshot(
-                ormsgpack.unpackb(
-                    data, ext_hook=ext_hook, option=ormsgpack.OPT_NON_STR_KEYS
-                )
-            )
-        elif code == EXT_CONSTRUCTOR_SINGLE_ARG:
+        if code == EXT_CONSTRUCTOR_SINGLE_ARG:
             try:
                 tup = ormsgpack.unpackb(
                     data, ext_hook=ext_hook, option=ormsgpack.OPT_NON_STR_KEYS

libs/checkpoint/langgraph/checkpoint/serde/types.py

@@ -1,7 +1,6 @@
 from collections.abc import Sequence
 from typing import (
     Any,
-    NamedTuple,
     Protocol,
     TypeVar,
     runtime_checkable,
@@ -34,20 +33,6 @@ class _DeltaSentinel:
 DELTA_SENTINEL = _DeltaSentinel()
 
 
-class _DeltaSnapshot(NamedTuple):
-    """Snapshot blob for a DeltaChannel with finite snapshot_frequency.
-
-    Stored in checkpoint_blobs via the `EXT_DELTA_SNAPSHOT` msgpack ext code.
-    The ancestor walk in `_get_channel_writes_history` terminates when it
-    encounters this type (any non-sentinel blob stops the walk).
-
-    `from_checkpoint` reconstructs the channel value directly from `.value`
-    without replaying writes — the snapshot IS the accumulated state.
-    """
-
-    value: Any
-
-
 Value = TypeVar("Value", covariant=True)
 Update = TypeVar("Update", contravariant=True)
 C = TypeVar("C")

libs/checkpoint/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "langgraph-checkpoint"
-version = "4.1.0a2"
+version = "4.0.2"
 description = "Library with base interfaces for LangGraph checkpoint savers."
 authors = []
 requires-python = ">=3.10"

libs/checkpoint/tests/test_jsonplus.py

@@ -1005,7 +1005,8 @@ def test_delta_sentinel_serde_round_trip() -> None:
 
     serde = JsonPlusSerializer()
     type_tag, blob = serde.dumps_typed(DELTA_SENTINEL)
-    assert type_tag == "msgpack"
-    assert blob  # non-empty ext envelope
+    # Zero-byte "delta" tag — no allowlist change needed.
+    assert type_tag == "delta"
+    assert blob == b""
     loaded = serde.loads_typed((type_tag, blob))
     assert loaded is DELTA_SENTINEL

libs/checkpoint/uv.lock

@@ -7,9 +7,6 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 
-[options]
-prerelease-mode = "allow"
-
 [[package]]
 name = "annotated-types"
 version = "0.7.0"
@@ -289,7 +286,7 @@ wheels = [
 
 [[package]]
 name = "langgraph-checkpoint"
-version = "4.1.0a2"
+version = "4.0.2"
 source = { editable = "." }
 dependencies = [
     { name = "langchain-core" },

libs/langgraph/langgraph/channels/__init__.py

@@ -1,7 +1,7 @@
+from langgraph.channels.aggregate import AggregateChannel
 from langgraph.channels.any_value import AnyValue
 from langgraph.channels.base import BaseChannel
 from langgraph.channels.binop import BinaryOperatorAggregate
-from langgraph.channels.delta import DeltaChannel
 from langgraph.channels.ephemeral_value import EphemeralValue
 from langgraph.channels.last_value import LastValue, LastValueAfterFinish
 from langgraph.channels.named_barrier_value import (
@@ -20,8 +20,8 @@ __all__ = (
     "LastValueAfterFinish",
     "UntrackedValue",
     "EphemeralValue",
+    "AggregateChannel",
     "BinaryOperatorAggregate",
-    "DeltaChannel",
     "NamedBarrierValue",
     "NamedBarrierValueAfterFinish",
     # topics

libs/langgraph/langgraph/channels/aggregate.py

@@ -0,0 +1,256 @@
+from __future__ import annotations
+
+import collections.abc
+import copy as _copy
+import math
+from collections.abc import Callable, Sequence
+from typing import Any, Generic
+
+from langgraph.checkpoint.base import DELTA_SENTINEL, PendingWrite
+from typing_extensions import NotRequired, Required, Self
+
+from langgraph._internal._constants import OVERWRITE
+from langgraph._internal._typing import MISSING
+from langgraph.channels.base import BaseChannel, Value
+from langgraph.errors import (
+    EmptyChannelError,
+    ErrorCode,
+    InvalidUpdateError,
+    create_error_message,
+)
+from langgraph.types import Overwrite
+
+__all__ = ("AggregateChannel",)
+
+
+def _strip_extras(t: Any) -> Any:
+    """Strips Annotated, Required, and NotRequired wrappers."""
+    if hasattr(t, "__origin__"):
+        return _strip_extras(t.__origin__)
+    if hasattr(t, "__origin__") and t.__origin__ in (Required, NotRequired):
+        return _strip_extras(t.__args__[0])
+    return t
+
+
+def _concrete(typ: Any) -> Any:
+    """Replace abstract collection types from `typing`/`collections.abc` with
+    their instantiable counterparts."""
+    typ = _strip_extras(typ)
+    if typ in (collections.abc.Sequence, collections.abc.MutableSequence):
+        return list
+    if typ in (collections.abc.Set, collections.abc.MutableSet):
+        return set
+    if typ in (collections.abc.Mapping, collections.abc.MutableMapping):
+        return dict
+    return typ
+
+
+def _empty(typ: Any) -> Any:
+    try:
+        return typ()
+    except Exception:
+        return []
+
+
+def _get_overwrite(value: Any) -> tuple[bool, Any]:
+    """Return (is_overwrite, overwrite_value) for an incoming write."""
+    if isinstance(value, Overwrite):
+        return True, value.value
+    if isinstance(value, dict) and set(value.keys()) == {OVERWRITE}:
+        return True, value[OVERWRITE]
+    return False, None
+
+
+class AggregateChannel(Generic[Value], BaseChannel[Value, Value, Any]):
+    """Fold-reducer channel with configurable snapshot cadence.
+
+    `snapshot_frequency=1` (default) writes a full blob every step — same
+    storage behavior as the classic `BinaryOperatorAggregate`.
+
+    `snapshot_frequency=N` (integer > 1) writes a sentinel on non-snapshot
+    steps; the value is reconstructed at read time by folding ancestor
+    writes through `operator`. On every Nth step a full blob is written,
+    bounding replay depth to N.
+
+    `snapshot_frequency=math.inf` never writes a blob — pure delta storage.
+    Reconstruction replays every write from thread start.
+
+    Parameters:
+        operator: Binary reducer `(Value, Value) -> Value` applied pairwise
+            to accumulate writes. Must be associative for correctness under
+            `snapshot_frequency != 1` where the fold order across ancestor
+            replay vs live writes differs from the classic single-fold path.
+            Most practical reducers (`operator.add`, `add_messages`) satisfy
+            this.
+        snapshot_frequency: Every Nth step writes a full snapshot blob.
+            Default 1 (snapshot always). `math.inf` for pure-delta mode.
+            Reading at step M with `snapshot_frequency=N` walks at most
+            `M % N` ancestor writes — bounded replay regardless of thread
+            depth.
+        typ: Value type. When used as an `Annotated[T, AggregateChannel(...)]`
+            state field, the type is inferred from `T` and this kwarg is
+            unused. Explicit kwarg is the escape hatch for imperative
+            graph construction.
+
+    Experimental under `snapshot_frequency > 1`: the sentinel+replay path
+    is the same mechanism that the (now removed) `DeltaChannel` used; the
+    cadence knob is new and should be validated on real workloads before
+    being relied on in production.
+    """
+
+    __slots__ = ("value", "operator", "snapshot_frequency", "_typ_provided")
+
+    def __init__(
+        self,
+        operator: Callable[[Value, Value], Value],
+        *,
+        snapshot_frequency: int | float = 1,
+        typ: type[Value] | None = None,
+    ) -> None:
+        self._typ_provided = typ is not None
+        concrete_typ = _concrete(typ) if typ is not None else list
+        super().__init__(concrete_typ)
+        self.operator = operator
+        self.snapshot_frequency = snapshot_frequency
+        try:
+            self.value = concrete_typ()
+        except Exception:
+            self.value = MISSING
+
+    def __eq__(self, other: object) -> bool:
+        if not isinstance(other, AggregateChannel):
+            return False
+        if self.snapshot_frequency != other.snapshot_frequency:
+            return False
+        if (
+            self.operator.__name__ != "<lambda>"
+            and other.operator.__name__ != "<lambda>"
+        ):
+            return self.operator is other.operator
+        return True
+
+    @property
+    def ValueType(self) -> Any:
+        return self.typ
+
+    @property
+    def UpdateType(self) -> Any:
+        return self.typ
+
+    def _clone_empty(self) -> Self:
+        """Create a blank clone preserving all attributes, bypassing __init__.
+
+        Subclasses (e.g. BinaryOperatorAggregate) have different __init__
+        signatures; going through __init__ from copy/from_checkpoint would
+        pass kwargs those subclasses don't accept. Bypassing avoids that.
+        """
+        new = self.__class__.__new__(self.__class__)
+        new.typ = self.typ
+        new.key = self.key
+        new.operator = self.operator
+        new.snapshot_frequency = self.snapshot_frequency
+        new._typ_provided = self._typ_provided
+        new.value = MISSING
+        return new
+
+    def copy(self) -> Self:
+        new = self._clone_empty()
+        new.value = self.value if self.value is MISSING else _copy.copy(self.value)
+        return new
+
+    def is_snapshot_step(self, step: int) -> bool:
+        """Return True if a full blob should be written at this step.
+
+        `snapshot_frequency=1` → always. `snapshot_frequency=math.inf` →
+        never. Otherwise, `step % snapshot_frequency == 0`.
+        """
+        if self.snapshot_frequency == 1:
+            return True
+        if self.snapshot_frequency == math.inf:
+            return False
+        return step % self.snapshot_frequency == 0
+
+    def _apply_write(self, value: Any, write: Any) -> Any:
+        """Apply one write and return the new value. Handles Overwrite."""
+        is_overwrite, overwrite_value = _get_overwrite(write)
+        if is_overwrite:
+            return (
+                _copy.copy(overwrite_value)
+                if overwrite_value is not None
+                else _empty(self.typ)
+            )
+        if value is MISSING:
+            return write
+        return self.operator(value, write)
+
+    def from_checkpoint(self, checkpoint: Any) -> Self:
+        """Initialize from a stored blob, sentinel, or MISSING.
+
+        If the stored value is a full blob, use it as-is. If it is
+        `DELTA_SENTINEL` or `MISSING`, start empty — the caller (pregel)
+        is responsible for replaying writes via `replay_writes` when
+        applicable.
+        """
+        new = self._clone_empty()
+        if checkpoint is MISSING or checkpoint is DELTA_SENTINEL:
+            new.value = _empty(self.typ)
+        else:
+            new.value = checkpoint
+        return new
+
+    def replay_writes(self, writes: Sequence[PendingWrite]) -> None:
+        """Fold a sequence of PendingWrite tuples into the current value.
+
+        Called by pregel after `from_checkpoint(seed)` to replay per-step
+        deltas from on-path ancestors through the operator. Writes are
+        oldest→newest. Overwrite markers reset the reducer state at that
+        point. `task_id` and `channel` fields are ignored — the caller
+        has already filtered to this channel.
+        """
+        for _, _, value in writes:
+            self.value = self._apply_write(self.value, value)
+
+    def update(self, values: Sequence[Value]) -> bool:
+        if not values:
+            return False
+        seen_overwrite = False
+        for value in values:
+            is_overwrite, _ = _get_overwrite(value)
+            if is_overwrite:
+                if seen_overwrite:
+                    msg = create_error_message(
+                        message="Can receive only one Overwrite value per super-step.",
+                        error_code=ErrorCode.INVALID_CONCURRENT_GRAPH_UPDATE,
+                    )
+                    raise InvalidUpdateError(msg)
+                seen_overwrite = True
+            elif seen_overwrite:
+                continue
+            self.value = self._apply_write(self.value, value)
+        return True
+
+    def get(self) -> Value:
+        if self.value is MISSING:
+            raise EmptyChannelError()
+        return self.value
+
+    def is_available(self) -> bool:
+        return self.value is not MISSING
+
+    def checkpoint(self) -> Any:
+        """Return the serializable representation of current state.
+
+        For `snapshot_frequency=math.inf` (pure delta), always returns
+        `DELTA_SENTINEL` — the value lives in `checkpoint_writes` and is
+        reconstructed by replay, never materialized as a blob.
+
+        For integer `snapshot_frequency`, returns the full value. Pregel's
+        `create_checkpoint` is responsible for consulting
+        `is_snapshot_step(step)` to decide whether to actually store the
+        full value or write `DELTA_SENTINEL` for that step.
+        """
+        if self.value is MISSING:
+            return MISSING
+        if self.snapshot_frequency == math.inf:
+            return DELTA_SENTINEL
+        return self.value

libs/langgraph/langgraph/channels/binop.py

@@ -1,54 +1,21 @@
-import collections.abc
-from collections.abc import Callable, Sequence
-from typing import Any, Generic
+from collections.abc import Callable
+from typing import Generic
 
-from typing_extensions import NotRequired, Required, Self
-
-from langgraph._internal._constants import OVERWRITE
-from langgraph._internal._typing import MISSING
-from langgraph.channels.base import BaseChannel, Value
-from langgraph.errors import (
-    EmptyChannelError,
-    ErrorCode,
-    InvalidUpdateError,
-    create_error_message,
+from langgraph.channels.aggregate import (
+    AggregateChannel,
 )
-from langgraph.types import Overwrite
+from langgraph.channels.aggregate import (
+    _get_overwrite as _get_overwrite,
+)
+from langgraph.channels.aggregate import (
+    _strip_extras as _strip_extras,
+)
+from langgraph.channels.base import Value
 
 __all__ = ("BinaryOperatorAggregate",)
 
 
-# Adapted from typing_extensions
-def _strip_extras(t):  # type: ignore[no-untyped-def]
-    """Strips Annotated, Required and NotRequired from a given type."""
-    if hasattr(t, "__origin__"):
-        if t.__origin__ in (Required, NotRequired):
-            return _strip_extras(t.__args__[0])
-        return _strip_extras(t.__origin__)
-    return t
-
-
-def _get_overwrite(value: Any) -> tuple[bool, Any]:
-    """Inspects the given value and returns (is_overwrite, overwrite_value)."""
-    if isinstance(value, Overwrite):
-        return True, value.value
-    if isinstance(value, dict) and len(value) == 1 and OVERWRITE in value:
-        return True, value[OVERWRITE]
-    return False, None
-
-
-def _operators_equal(a: Callable, b: Callable) -> bool:
-    """Return True if two reducer operators should be considered equal.
-
-    Lambdas all share the name '<lambda>' so identity comparison is
-    unreliable; treat any pairing that includes a lambda as equal.
-    """
-    if a.__name__ == "<lambda>" or b.__name__ == "<lambda>":
-        return True
-    return a is b
-
-
-class BinaryOperatorAggregate(Generic[Value], BaseChannel[Value, Value, Value]):
+class BinaryOperatorAggregate(AggregateChannel[Value], Generic[Value]):
     """Stores the result of applying a binary operator to the current value and each new value.
 
     ```python
@@ -56,86 +23,21 @@ class BinaryOperatorAggregate(Generic[Value], BaseChannel[Value, Value, Value]):
 
     total = Channels.BinaryOperatorAggregate(int, operator.add)
     ```
+
+    Equivalent to `AggregateChannel(operator, typ=typ, snapshot_frequency=1)`.
+    Preserved as a distinct subclass so existing `isinstance(x, BinaryOperatorAggregate)`
+    checks and `_is_field_binop` detection continue to work. New code should
+    prefer `AggregateChannel` directly — especially when a non-unit
+    `snapshot_frequency` is wanted.
     """
 
-    __slots__ = ("value", "operator")
-
     def __init__(self, typ: type[Value], operator: Callable[[Value, Value], Value]):
-        super().__init__(typ)
-        self.operator = operator
-        # special forms from typing or collections.abc are not instantiable
-        # so we need to replace them with their concrete counterparts
-        typ = _strip_extras(typ)
-        if typ in (collections.abc.Sequence, collections.abc.MutableSequence):
-            typ = list
-        if typ in (collections.abc.Set, collections.abc.MutableSet):
-            typ = set
-        if typ in (collections.abc.Mapping, collections.abc.MutableMapping):
-            typ = dict
-        try:
-            self.value = typ()
-        except Exception:
-            self.value = MISSING
+        super().__init__(operator, typ=typ, snapshot_frequency=1)
 
     def __eq__(self, value: object) -> bool:
-        return isinstance(value, BinaryOperatorAggregate) and _operators_equal(
-            self.operator, value.operator
+        return isinstance(value, BinaryOperatorAggregate) and (
+            value.operator is self.operator
+            if value.operator.__name__ != "<lambda>"
+            and self.operator.__name__ != "<lambda>"
+            else True
         )
-
-    @property
-    def ValueType(self) -> type[Value]:
-        """The type of the value stored in the channel."""
-        return self.typ
-
-    @property
-    def UpdateType(self) -> type[Value]:
-        """The type of the update received by the channel."""
-        return self.typ
-
-    def copy(self) -> Self:
-        """Return a copy of the channel."""
-        empty = self.__class__(self.typ, self.operator)
-        empty.key = self.key
-        empty.value = self.value
-        return empty
-
-    def from_checkpoint(self, checkpoint: Value) -> Self:
-        empty = self.__class__(self.typ, self.operator)
-        empty.key = self.key
-        if checkpoint is not MISSING:
-            empty.value = checkpoint
-        return empty
-
-    def update(self, values: Sequence[Value]) -> bool:
-        if not values:
-            return False
-        if self.value is MISSING:
-            self.value = values[0]
-            values = values[1:]
-        seen_overwrite: bool = False
-        for value in values:
-            is_overwrite, overwrite_value = _get_overwrite(value)
-            if is_overwrite:
-                if seen_overwrite:
-                    msg = create_error_message(
-                        message="Can receive only one Overwrite value per super-step.",
-                        error_code=ErrorCode.INVALID_CONCURRENT_GRAPH_UPDATE,
-                    )
-                    raise InvalidUpdateError(msg)
-                self.value = overwrite_value
-                seen_overwrite = True
-                continue
-            if not seen_overwrite:
-                self.value = self.operator(self.value, value)
-        return True
-
-    def get(self) -> Value:
-        if self.value is MISSING:
-            raise EmptyChannelError()
-        return self.value
-
-    def is_available(self) -> bool:
-        return self.value is not MISSING
-
-    def checkpoint(self) -> Value:
-        return self.value

libs/langgraph/langgraph/channels/delta.py

@@ -1,197 +0,0 @@
-from __future__ import annotations
-
-import collections.abc
-import copy as _copy
-from collections.abc import Callable, Sequence
-from typing import Any, Generic
-
-from langgraph.checkpoint.base import DELTA_SENTINEL, PendingWrite
-from langgraph.checkpoint.serde.types import _DeltaSnapshot
-from typing_extensions import Self
-
-from langgraph._internal._typing import MISSING
-from langgraph.channels.base import BaseChannel, Value
-from langgraph.channels.binop import _get_overwrite, _operators_equal, _strip_extras
-from langgraph.errors import (
-    EmptyChannelError,
-    ErrorCode,
-    InvalidUpdateError,
-    create_error_message,
-)
-
-__all__ = ("DeltaChannel",)
-
-
-class DeltaChannel(Generic[Value], BaseChannel[Any, Any, Any]):
-    """Reducer channel that stores only a sentinel in checkpoint blobs and
-    reconstructs state by replaying ancestor writes through the reducer.
-
-    The reducer receives the current accumulated value and a batch of writes
-    in one call: `reducer(state, [write1, write2, ...]) -> new_state`.
-
-    Reducers must be deterministic and batching-invariant (associative across
-    folds): applying two consecutive write batches separately must produce the
-    same state as applying their concatenation once:
-
-        reducer(reducer(state, xs), ys) == reducer(state, xs + ys)
-
-    This lets LangGraph replay checkpointed writes in larger batches than they
-    were originally produced without changing reconstructed state.
-
-    `snapshot_frequency=None` (default): pure delta; stores only
-    `DELTA_SENTINEL` in checkpoint blobs; reads replay all ancestor writes.
-
-    `snapshot_frequency=N`: `create_checkpoint` writes a full `_DeltaSnapshot`
-    blob every N steps, bounding replay depth to N.
-
-    Parameters:
-        reducer: `(state, list[writes]) -> new_state`. Must be deterministic
-            and batching-invariant as described above.
-        typ: The value type (e.g. `list`, `dict`). Inferred automatically
-            from the outer type when used inside `Annotated[T, DeltaChannel(...)]`.
-        snapshot_frequency: Every Nth pregel step writes a snapshot blob.
-            `None` (default) = pure delta, never snapshot.
-    """
-
-    __slots__ = ("value", "reducer", "snapshot_frequency")
-    value: Value | Any
-
-    def __init__(
-        self,
-        reducer: Callable[[Any, Sequence[Any]], Any],
-        typ: type[Value] | None = None,
-        *,
-        snapshot_frequency: int | None = None,
-    ) -> None:
-        if typ is None:
-            typ = list  # type: ignore[assignment]  # placeholder; overridden by _is_field_channel
-        super().__init__(typ)
-        self.reducer = reducer
-        self.snapshot_frequency = snapshot_frequency
-        typ = _strip_extras(typ)
-        if typ in (collections.abc.Sequence, collections.abc.MutableSequence):
-            typ = list
-        if typ in (collections.abc.Set, collections.abc.MutableSet):
-            typ = set
-        if typ in (collections.abc.Mapping, collections.abc.MutableMapping):
-            typ = dict
-        self.typ = typ
-        self.value: Any = MISSING
-
-    def __eq__(self, other: object) -> bool:
-        if not isinstance(other, DeltaChannel):
-            return False
-        if self.snapshot_frequency != other.snapshot_frequency:
-            return False
-        return _operators_equal(self.reducer, other.reducer)
-
-    @property
-    def ValueType(self) -> Any:
-        return self.typ
-
-    @property
-    def UpdateType(self) -> Any:
-        return self.typ
-
-    def is_snapshot_step(self, step: int) -> bool:
-        """True if pregel should write a snapshot blob at this step."""
-        return (
-            self.snapshot_frequency is not None
-            and step > 0
-            and step % self.snapshot_frequency == 0
-        )
-
-    def copy(self) -> Self:
-        new = self.__class__(
-            self.reducer, self.typ, snapshot_frequency=self.snapshot_frequency
-        )
-        new.key = self.key
-        new.value = self.value if self.value is MISSING else _copy.copy(self.value)
-        return new
-
-    def from_checkpoint(self, checkpoint: Any) -> Self:
-        """Initialize from a stored blob or sentinel.
-
-        Blob types (dispatched via serde ext code, not dict key inspection):
-          * `DELTA_SENTINEL` / `MISSING`: start empty; caller replays writes.
-          * `_DeltaSnapshot(value)`: restore value directly from snapshot.
-          * plain value (migration from old BinOp blobs): use directly.
-        """
-        new = self.__class__(
-            self.reducer, self.typ, snapshot_frequency=self.snapshot_frequency
-        )
-        new.key = self.key
-        if checkpoint is MISSING or checkpoint is DELTA_SENTINEL:
-            new.value = self.typ()
-        elif isinstance(checkpoint, _DeltaSnapshot):
-            new.value = checkpoint.value
-        else:
-            new.value = checkpoint
-        return new
-
-    def replay_writes(self, writes: Sequence[PendingWrite]) -> None:
-        """Apply ancestor writes oldest-to-newest via a single reducer call.
-
-        If any write is an Overwrite, the last one in the sequence acts as
-        the reset point: its value becomes the new base and only writes
-        after it are passed to the reducer.
-        """
-        values = [v for _, _, v in writes]
-        if not values:
-            return
-        base = self.value
-        start = 0
-        for i, v in enumerate(values):
-            is_ow, ow_value = _get_overwrite(v)
-            if is_ow:
-                base = _copy.copy(ow_value) if ow_value is not None else self.typ()
-                start = i + 1
-        remaining = values[start:]
-        self.value = self.reducer(base, remaining) if remaining else base
-
-    def update(self, values: Sequence[Any]) -> bool:
-        if not values:
-            return False
-        overwrite_idx: int | None = None
-        for i, v in enumerate(values):
-            is_ow, _ = _get_overwrite(v)
-            if is_ow:
-                if overwrite_idx is not None:
-                    msg = create_error_message(
-                        message="Can receive only one Overwrite value per super-step.",
-                        error_code=ErrorCode.INVALID_CONCURRENT_GRAPH_UPDATE,
-                    )
-                    raise InvalidUpdateError(msg)
-                overwrite_idx = i
-        if overwrite_idx is not None:
-            _, overwrite_value = _get_overwrite(values[overwrite_idx])
-            base = (
-                _copy.copy(overwrite_value)
-                if overwrite_value is not None
-                else self.typ()
-            )
-            remaining = [v for i, v in enumerate(values) if i != overwrite_idx]
-            self.value = self.reducer(base, remaining) if remaining else base
-            return True
-        base = self.typ() if self.value is MISSING else self.value
-        self.value = self.reducer(base, list(values))
-        return True
-
-    def get(self) -> Any:
-        if self.value is MISSING:
-            raise EmptyChannelError()
-        return self.value
-
-    def is_available(self) -> bool:
-        return self.value is not MISSING
-
-    def checkpoint(self) -> Any:
-        """Return stored representation: always `DELTA_SENTINEL`.
-
-        Snapshot decisions are made by `create_checkpoint` in pregel (which
-        has the step number) via `is_snapshot_step`. `checkpoint()` is only
-        called for non-snapshot steps or when no checkpointer is available.
-        """
-        if self.value is MISSING:
-            return MISSING
-        return DELTA_SENTINEL

libs/langgraph/langgraph/graph/message.py

@@ -244,52 +244,6 @@ def add_messages(
     return merged
 
 
-def _messages_delta_reducer(
-    state: list[AnyMessage], writes: list[list[AnyMessage]]
-) -> list[AnyMessage]:
-    """**Experimental.** Batch reducer for use with `DeltaChannel`.
-
-    Processes all writes in one pass — dedup by ID, `RemoveMessage`
-    tombstoning — without calling `add_messages`. Assumes writes contain
-    already-typed `BaseMessage` objects (no raw-dict coercion).
-
-    This reducer is batching-invariant, as required by `DeltaChannel`:
-    `reducer(reducer(state, xs), ys) == reducer(state, xs + ys)`.
-
-    Use `add_messages` as the reducer for `BinaryOperatorAggregate` or
-    anywhere raw message dicts / strings need to be coerced first.
-
-    Example::
-
-        from typing import Annotated
-        from langgraph.channels.delta import DeltaChannel
-        from langgraph.graph.message import _messages_delta_reducer
-
-        class State(TypedDict):
-            messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
-    """
-    from itertools import chain
-
-    index: dict[str, int] = {m.id: i for i, m in enumerate(state) if m.id is not None}
-    result: list[AnyMessage | None] = list(state)
-    for msg in chain.from_iterable(
-        [w] if isinstance(w, BaseMessage) else w for w in writes
-    ):
-        mid = msg.id
-        if mid is None:
-            result.append(msg)
-        elif isinstance(msg, RemoveMessage):
-            if mid in index:
-                result[index[mid]] = None
-                del index[mid]
-        elif mid in index:
-            result[index[mid]] = msg
-        else:
-            index[mid] = len(result)
-            result.append(msg)
-    return [m for m in result if m is not None]
-
-
 @deprecated(
     "MessageGraph is deprecated in langgraph 1.0.0, to be removed in 2.0.0. Please use StateGraph with a `messages` key instead.",
     category=None,

libs/langgraph/langgraph/graph/state.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+import collections.abc
 import inspect
 import logging
 import typing
@@ -46,9 +47,9 @@ from langgraph._internal._fields import (
 from langgraph._internal._pydantic import create_model
 from langgraph._internal._runnable import coerce_to_runnable
 from langgraph._internal._typing import EMPTY_SEQ, MISSING, DeprecatedKwargs
+from langgraph.channels.aggregate import AggregateChannel
 from langgraph.channels.base import BaseChannel
-from langgraph.channels.binop import BinaryOperatorAggregate
-from langgraph.channels.delta import DeltaChannel
+from langgraph.channels.binop import BinaryOperatorAggregate, _strip_extras
 from langgraph.channels.ephemeral_value import EphemeralValue
 from langgraph.channels.last_value import LastValue, LastValueAfterFinish
 from langgraph.channels.named_barrier_value import (
@@ -1669,7 +1670,15 @@ def _is_field_channel(typ: type[Any]) -> BaseChannel | None:
         # Search through all annotated medata to find channel annotations
         for item in meta:
             if isinstance(item, BaseChannel):
-                if isinstance(item, DeltaChannel) and hasattr(typ, "__origin__"):
+                # AggregateChannel instances without an explicit `typ` arg
+                # inherit the value type from the outer `Annotated[...]`.
+                # BinaryOperatorAggregate (a subclass) always sets typ
+                # explicitly, so _typ_provided is True and we skip inference.
+                if (
+                    isinstance(item, AggregateChannel)
+                    and not item._typ_provided
+                    and hasattr(typ, "__origin__")
+                ):
                     origin = typ.__origin__
                     # Unwrap parameterized Required[X]/NotRequired[X] to X
                     # (e.g. Annotated[NotRequired[dict[...]], ...]).
@@ -1678,11 +1687,27 @@ def _is_field_channel(typ: type[Any]) -> BaseChannel | None:
                         NotRequired,
                     ):
                         origin = origin.__args__[0]
-                    item = item.__class__(
-                        item.reducer,
-                        origin,
-                        snapshot_frequency=item.snapshot_frequency,
-                    )
+                    outer = _strip_extras(origin)
+                    if outer in (
+                        collections.abc.Sequence,
+                        collections.abc.MutableSequence,
+                    ):
+                        outer = list
+                    elif outer in (
+                        collections.abc.Mapping,
+                        collections.abc.MutableMapping,
+                    ):
+                        outer = dict
+                    elif outer in (
+                        collections.abc.Set,
+                        collections.abc.MutableSet,
+                    ):
+                        outer = set
+                    item.typ = outer
+                    try:
+                        item.value = outer()
+                    except Exception:
+                        item.value = []
                 return item
             elif isclass(item) and issubclass(item, BaseChannel):
                 # ex, Annotated[int, EphemeralValue, SomeOtherAnnotation]

libs/langgraph/langgraph/pregel/_checkpoint.py

@@ -1,23 +1,19 @@
 from __future__ import annotations
 
-from collections.abc import Callable, Mapping
+from collections.abc import Mapping
 from datetime import datetime, timezone
-from typing import Any, cast
 
 from langchain_core.runnables import RunnableConfig
 from langgraph.checkpoint.base import DELTA_SENTINEL, BaseCheckpointSaver, Checkpoint
 from langgraph.checkpoint.base.id import uuid6
-from langgraph.checkpoint.serde.types import _DeltaSnapshot
 
 from langgraph._internal._typing import MISSING
+from langgraph.channels.aggregate import AggregateChannel
 from langgraph.channels.base import BaseChannel
-from langgraph.channels.delta import DeltaChannel
 from langgraph.managed.base import ManagedValueMapping, ManagedValueSpec
 
 LATEST_VERSION = 4
 
-GetNextVersion = Callable[[Any, None], Any]
-
 
 def empty_checkpoint() -> Checkpoint:
     return Checkpoint(
@@ -37,68 +33,46 @@ def create_checkpoint(
     *,
     id: str | None = None,
     updated_channels: set[str] | None = None,
-    get_next_version: GetNextVersion | None = None,
-    force_delta_snapshot: bool = False,
 ) -> Checkpoint:
     """Create a checkpoint for the given channels.
 
-    For `DeltaChannel` with `snapshot_frequency=N`, snapshot steps write a
-    `_DeltaSnapshot` blob rather than `DELTA_SENTINEL`, bounding the ancestor
-    walk to at most N steps. Snapshots are eager: even if the channel had no
-    write this step, a version bump is forced (via `get_next_version`) so the
-    blob is stored by `put()`. Without `get_next_version` (e.g. static
-    contexts), snapshot steps gracefully fall back to sentinel.
-
-    `force_delta_snapshot` writes available `DeltaChannel` values as snapshots
-    regardless of `snapshot_frequency`. This is used by `durability="exit"`,
-    where intermediate writes are not stored as ancestor `checkpoint_writes`.
+    For `AggregateChannel` spec with `snapshot_frequency != 1`, the stored
+    blob alternates between the full value and `DELTA_SENTINEL` based on
+    `is_snapshot_step(step)`. Non-snapshot steps store the sentinel; the
+    value is reconstructed from ancestor writes at read time.
     """
     ts = datetime.now(timezone.utc).isoformat()
     if channels is None:
         values = checkpoint["channel_values"]
-        channel_versions = checkpoint["channel_versions"]
     else:
         values = {}
-        channel_versions = dict(checkpoint["channel_versions"])
         for k in channels:
-            if k not in channel_versions:
+            if k not in checkpoint["channel_versions"]:
                 continue
             ch = channels[k]
-            if (
-                isinstance(ch, DeltaChannel)
-                and (force_delta_snapshot or ch.is_snapshot_step(step))
-                and ch.is_available()
-            ):
-                # Eager snapshot: bump version if not already written this step
-                # so put() includes this channel in new_versions and stores blob.
-                if get_next_version is not None and (
-                    updated_channels is None or k not in updated_channels
-                ):
-                    channel_versions[k] = get_next_version(channel_versions[k], None)
-                values[k] = _DeltaSnapshot(ch.get())
-            else:
-                v = ch.checkpoint()
-                if v is not MISSING:
-                    values[k] = v
+            if isinstance(ch, AggregateChannel) and not ch.is_snapshot_step(step):
+                values[k] = DELTA_SENTINEL
+                continue
+            v = ch.checkpoint()
+            if v is not MISSING:
+                values[k] = v
     return Checkpoint(
         v=LATEST_VERSION,
         ts=ts,
         id=id or str(uuid6(clock_seq=step)),
         channel_values=values,
-        channel_versions=channel_versions,
+        channel_versions=checkpoint["channel_versions"],
         versions_seen=checkpoint["versions_seen"],
         updated_channels=None if updated_channels is None else sorted(updated_channels),
     )
 
 
 def _needs_replay(spec: BaseChannel, stored: object) -> bool:
-    """True if `spec` is a `DeltaChannel` and the stored blob is a sentinel,
-    requiring an ancestor walk to reconstruct.
-
-    `_DeltaSnapshot` blobs and plain values (migration) resolve directly via
-    `from_checkpoint` — only `DELTA_SENTINEL` / `MISSING` trigger replay.
-    """
-    if not isinstance(spec, DeltaChannel):
+    """True if `spec` is a delta-mode AggregateChannel and the stored
+    blob is empty/sentinel, requiring an ancestor walk to reconstruct."""
+    if not isinstance(spec, AggregateChannel):
+        return False
+    if spec.snapshot_frequency == 1:
         return False
     return stored is MISSING or stored is DELTA_SENTINEL
 
@@ -113,10 +87,15 @@ def channels_from_checkpoint(
     """Hydrate channels from a checkpoint.
 
     For most channels, `spec.from_checkpoint(checkpoint["channel_values"][k])`
-    is sufficient. `DeltaChannel` is the exception: sentinel blobs require an
-    ancestor walk via `saver._get_channel_writes_history`. The walk terminates
-    at the nearest `_DeltaSnapshot` blob (step-based) or a pre-migration plain
-    value, so read depth is bounded by `snapshot_frequency`.
+    is sufficient — the stored value IS the reconstructed state.
+
+    `AggregateChannel` with `snapshot_frequency != 1` is the exception:
+    its stored value on non-snapshot steps is `DELTA_SENTINEL`; the full
+    state is spread across `checkpoint_writes` along the ancestor chain.
+    When `saver` and `config` are provided, this function fetches that
+    history via `saver._get_channel_writes_history` and folds it through
+    the channel's operator. Without them (static contexts — graph
+    drawing, unit tests), delta-mode channels fall back to empty.
     """
     channel_specs: dict[str, BaseChannel] = {}
     managed_specs: dict[str, ManagedValueSpec] = {}
@@ -131,9 +110,12 @@ def channels_from_checkpoint(
         ch: BaseChannel
         stored = checkpoint["channel_values"].get(k, MISSING)
         if _needs_replay(spec, stored) and saver is not None and config is not None:
-            delta_spec = cast(DeltaChannel, spec)
+            # Walk ancestors for seed + writes. The saver's walk stops at
+            # the nearest non-sentinel blob (natural terminator under
+            # snapshot_frequency > 1; pre-migration blobs also act as
+            # terminators if the spec was changed mid-thread).
             history = saver._get_channel_writes_history(config, k)
-            replay_ch = delta_spec.from_checkpoint(history.seed)
+            replay_ch = spec.from_checkpoint(history.seed)
             replay_ch.replay_writes(history.writes)
             ch = replay_ch
         else:
@@ -163,9 +145,8 @@ async def achannels_from_checkpoint(
         ch: BaseChannel
         stored = checkpoint["channel_values"].get(k, MISSING)
         if _needs_replay(spec, stored) and saver is not None and config is not None:
-            delta_spec = cast(DeltaChannel, spec)
             history = await saver._aget_channel_writes_history(config, k)
-            replay_ch = delta_spec.from_checkpoint(history.seed)
+            replay_ch = spec.from_checkpoint(history.seed)
             replay_ch.replay_writes(history.writes)
             ch = replay_ch
         else:

libs/langgraph/langgraph/pregel/_loop.py

@@ -68,7 +68,6 @@ from langgraph.callbacks import (
     GraphResumeEvent,
 )
 from langgraph.channels.base import BaseChannel
-from langgraph.channels.delta import DeltaChannel
 from langgraph.channels.untracked_value import UntrackedValue
 from langgraph.constants import TAG_HIDDEN
 from langgraph.errors import (
@@ -190,8 +189,6 @@ class PregelLoop:
     _migrate_checkpoint: Callable[[Checkpoint], None] | None
     submit: Submit
     channels: Mapping[str, BaseChannel]
-    # Only set on AsyncPregelLoop; sync loops keep this as None.
-    _delta_write_futs: list[Any] | None = None
     managed: ManagedValueMapping
     checkpoint: Checkpoint
     checkpoint_id_saved: str
@@ -410,7 +407,7 @@ class PregelLoop:
                     task = self.tasks.get(task_id)
                 else:
                     task = None
-                fut = self.submit(
+                self.submit(
                     self.checkpointer_put_writes,
                     config,
                     writes_to_save,
@@ -418,16 +415,12 @@ class PregelLoop:
                     task_path_str(task.path) if task else "",
                 )
             else:
-                fut = self.submit(
+                self.submit(
                     self.checkpointer_put_writes,
                     config,
                     writes_to_save,
                     task_id,
                 )
-            if self._delta_write_futs is not None and any(
-                isinstance(self.specs.get(c), DeltaChannel) for c, _ in writes_to_save
-            ):
-                self._delta_write_futs.append(fut)
         # output writes
         if hasattr(self, "tasks"):
             self.output_writes(task_id, writes)
@@ -898,10 +891,6 @@ class PregelLoop:
             self.step,
             id=self.checkpoint["id"] if exiting else None,
             updated_channels=self.updated_channels,
-            get_next_version=self.checkpointer_get_next_version
-            if do_checkpoint
-            else None,
-            force_delta_snapshot=exiting and self.durability == "exit",
         )
         # sanitize TASK channel in the checkpoint before saving (durability=="exit")
         if TASKS in self.checkpoint["channel_values"] and any(
@@ -1383,11 +1372,6 @@ class AsyncPregelLoop(PregelLoop, AbstractAsyncContextManager):
         metadata: CheckpointMetadata,
         new_versions: ChannelVersions,
     ) -> RunnableConfig:
-        # Drain DeltaChannel write futures before committing the checkpoint so
-        # DELTA_SENTINEL blobs are never saved ahead of their backing writes.
-        if self._delta_write_futs:
-            futs, self._delta_write_futs = self._delta_write_futs, []
-            await asyncio.gather(*futs)
         try:
             if prev is not None:
                 await prev
@@ -1493,7 +1477,6 @@ class AsyncPregelLoop(PregelLoop, AbstractAsyncContextManager):
             if saved.pending_writes is not None
             else []
         )
-        self._delta_write_futs = []
         self.submit = await self.stack.enter_async_context(
             AsyncBackgroundExecutor(self.config)
         )

libs/langgraph/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "langgraph"
-version = "1.2.0a2"
+version = "1.1.9"
 description = "Building stateful, multi-actor applications with LLMs"
 authors = []
 requires-python = ">=3.10"
@@ -24,8 +24,8 @@ classifiers = [
     'Programming Language :: Python :: 3.13',
 ]
 dependencies = [
-    "langchain-core>=1.3.2,<2",
-    "langgraph-checkpoint>=4.0.3,<5.0.0",
+    "langchain-core>=1.3.0,<2",
+    "langgraph-checkpoint>=2.1.0,<5.0.0",
     "langgraph-sdk>=0.3.0,<0.4.0",
     "langgraph-prebuilt>=1.0.9,<1.1.0",
     "xxhash>=3.5.0",

libs/langgraph/tests/test_channels.py

@@ -1,34 +1,27 @@
 import operator
 from collections.abc import Sequence
-from typing import Annotated
 
 import pytest
-from langchain_core.messages import AIMessage, HumanMessage, RemoveMessage
+from langchain_core.messages import AIMessage, HumanMessage
 from langgraph.checkpoint.base import DELTA_SENTINEL
-from langgraph.checkpoint.memory import InMemorySaver
-from langgraph.checkpoint.serde.types import _DeltaSnapshot
-from typing_extensions import NotRequired, TypedDict
 
 from langgraph._internal._typing import MISSING
 from langgraph.channels.binop import BinaryOperatorAggregate
-from langgraph.channels.delta import DeltaChannel
 from langgraph.channels.last_value import LastValue
 from langgraph.channels.topic import Topic
 from langgraph.channels.untracked_value import UntrackedValue
 from langgraph.errors import EmptyChannelError, InvalidUpdateError
-from langgraph.graph import START, StateGraph
-from langgraph.graph.message import _messages_delta_reducer
-from langgraph.graph.state import _get_channel
-from langgraph.types import Overwrite
+from langgraph.graph.message import add_messages
+
+import math as _math_compat
+from langgraph.channels.aggregate import AggregateChannel as _AggregateChannel_compat
+def DeltaChannel(op):
+    return _AggregateChannel_compat(op, snapshot_frequency=_math_compat.inf)
+
 
 pytestmark = pytest.mark.anyio
 
 
-# ---------------------------------------------------------------------------
-# Core channel primitives
-# ---------------------------------------------------------------------------
-
-
 def test_last_value() -> None:
     channel = LastValue(int).from_checkpoint(MISSING)
     assert channel.ValueType is int
@@ -111,41 +104,49 @@ def test_untracked_value() -> None:
     assert channel.ValueType is dict
     assert channel.UpdateType is dict
 
+    # UntrackedValue should start empty
     with pytest.raises(EmptyChannelError):
         channel.get()
 
+    # Should be able to update with a value
     test_data = {"session": "test", "temp": "dir"}
     channel.update([test_data])
     assert channel.get() == test_data
 
+    # Update with new value
     new_data = {"session": "updated", "temp": "newdir"}
     channel.update([new_data])
     assert channel.get() == new_data
 
+    # On checkpoint, UntrackedValue should return MISSING
     checkpoint = channel.checkpoint()
     assert checkpoint is MISSING
 
+    # Creating from checkpoint with MISSING should start empty
     new_channel = UntrackedValue(dict).from_checkpoint(checkpoint)
     with pytest.raises(EmptyChannelError):
         new_channel.get()
 
 
-# ---------------------------------------------------------------------------
-# DeltaChannel — message reducer
-# ---------------------------------------------------------------------------
-
-
 def test_delta_channel_basic_two_steps() -> None:
-    ch = DeltaChannel(_messages_delta_reducer, list).from_checkpoint(MISSING)
+    from langchain_core.messages import AIMessage, HumanMessage
+    from langgraph.checkpoint.base import DELTA_SENTINEL
 
+    from langgraph.graph.message import add_messages
+
+    ch = DeltaChannel(add_messages).from_checkpoint(MISSING)
+
+    # Step 1: one message added
     ch.update([HumanMessage(content="hi", id="h1")])
     d1 = ch.checkpoint()
     assert d1 is DELTA_SENTINEL
 
+    # Step 2: another message
     ch.update([AIMessage(content="hello", id="a1")])
     d2 = ch.checkpoint()
     assert d2 is DELTA_SENTINEL
 
+    # Full accumulated value is preserved in memory
     assert len(ch.get()) == 2
     assert ch.get()[0].content == "hi"
     assert ch.get()[1].content == "hello"
@@ -153,7 +154,11 @@ def test_delta_channel_basic_two_steps() -> None:
 
 def test_delta_channel_from_checkpoint_writes_list() -> None:
     """replay_writes on a fresh channel replays through the operator."""
-    spec = DeltaChannel(_messages_delta_reducer, list)
+    from langchain_core.messages import AIMessage, HumanMessage
+
+    from langgraph.graph.message import add_messages
+
+    spec = DeltaChannel(add_messages)
     ch = spec.from_checkpoint(DELTA_SENTINEL)
     ch.replay_writes(
         [
@@ -170,28 +175,45 @@ def test_delta_channel_from_checkpoint_writes_list() -> None:
 
 
 def test_delta_channel_from_checkpoint_backwards_compat() -> None:
-    spec = DeltaChannel(_messages_delta_reducer, list)
+    from langchain_core.messages import HumanMessage
+
+    from langgraph.graph.message import add_messages
+
+    # Old BinaryOperatorAggregate checkpoint: plain list treated as backward compat
+    spec = DeltaChannel(add_messages)
     old_value = [HumanMessage(content="old", id="h1")]
     ch = spec.from_checkpoint(old_value)
     assert ch.get() == old_value
 
 
 def test_delta_channel_overwrite() -> None:
-    ch = DeltaChannel(_messages_delta_reducer, list).from_checkpoint(MISSING)
+    from langchain_core.messages import HumanMessage
+    from langgraph.checkpoint.base import DELTA_SENTINEL
+
+    from langgraph.graph.message import add_messages
+    from langgraph.types import Overwrite
+
+    ch = DeltaChannel(add_messages).from_checkpoint(MISSING)
     ch.update([HumanMessage(content="old", id="h1")])
 
     ch.update([Overwrite([HumanMessage(content="new", id="h2")])])
     d = ch.checkpoint()
     assert d is DELTA_SENTINEL
+    # After overwrite, value is reset to only the new message
     assert len(ch.get()) == 1
     assert ch.get()[0].content == "new"
 
 
 def test_delta_channel_remove_message_and_replay() -> None:
     """RemoveMessage must round-trip correctly when writes are replayed."""
-    spec = DeltaChannel(_messages_delta_reducer, list)
+    from langchain_core.messages import AIMessage, HumanMessage, RemoveMessage
+
+    from langgraph.graph.message import add_messages
+
+    spec = DeltaChannel(add_messages)
     ch = spec.from_checkpoint(MISSING)
 
+    # Step 1: add two messages
     ch.update([HumanMessage(content="hi", id="h1")])
     ch.update([AIMessage(content="hello", id="a1")])
     assert ch.get() == [
@@ -199,9 +221,11 @@ def test_delta_channel_remove_message_and_replay() -> None:
         AIMessage(content="hello", id="a1"),
     ]
 
+    # Step 2: remove the AI message
     ch.update([RemoveMessage(id="a1")])
     assert ch.get() == [HumanMessage(content="hi", id="h1")]
 
+    # Replay the writes list from scratch — must reproduce the post-remove state
     ch2 = spec.from_checkpoint(DELTA_SENTINEL)
     ch2.replay_writes(
         [
@@ -215,13 +239,21 @@ def test_delta_channel_remove_message_and_replay() -> None:
 
 def test_delta_channel_update_by_id_and_replay() -> None:
     """Updating a message by ID must round-trip correctly through writes replay."""
-    spec = DeltaChannel(_messages_delta_reducer, list)
+    from langchain_core.messages import HumanMessage
+
+    from langgraph.graph.message import add_messages
+
+    spec = DeltaChannel(add_messages)
     ch = spec.from_checkpoint(MISSING)
 
+    # Step 1: add a message
     ch.update([HumanMessage(content="original", id="h1")])
+
+    # Step 2: update the same message by ID
     ch.update([HumanMessage(content="updated", id="h1")])
     assert ch.get() == [HumanMessage(content="updated", id="h1")]
 
+    # Replay writes — must produce the updated message, not the original
     ch2 = spec.from_checkpoint(DELTA_SENTINEL)
     ch2.replay_writes(
         [
@@ -235,123 +267,32 @@ def test_delta_channel_update_by_id_and_replay() -> None:
 
 def test_delta_channel_checkpoint_returns_sentinel() -> None:
     """checkpoint() always returns DELTA_SENTINEL regardless of state."""
-    ch = DeltaChannel(_messages_delta_reducer, list).from_checkpoint(MISSING)
+    from langgraph.checkpoint.base import DELTA_SENTINEL
+
+    from langgraph.graph.message import add_messages
+
+    ch = DeltaChannel(add_messages).from_checkpoint(MISSING)
     assert ch.checkpoint() is DELTA_SENTINEL
 
+    from langchain_core.messages import HumanMessage
+
     ch.update([HumanMessage(content="hi", id="h1")])
     assert ch.checkpoint() is DELTA_SENTINEL
 
 
-# ---------------------------------------------------------------------------
-# DeltaChannel — snapshot frequency
-# ---------------------------------------------------------------------------
-
-
-def test_delta_channel_snapshot_step_based() -> None:
-    """Snapshots fire on every Nth step regardless of whether the channel was written.
-
-    With snapshot_frequency=N, every Nth pregel step produces a _DeltaSnapshot
-    blob — even if the channel had no write that step (eager snapshot). This
-    bounds the ancestor walk to at most N steps on any read.
-    """
-
-    class State(TypedDict):
-        messages: Annotated[
-            list, DeltaChannel(_messages_delta_reducer, snapshot_frequency=5)
-        ]
-        other: str
-
-    def node_a(state: State) -> dict:
-        i = len(state["messages"]) // 2
-        return {"messages": [AIMessage(content=f"a{i}", id=f"a{i}")]}
-
-    def node_b(state: State) -> dict:
-        return {"other": "y"}
-
-    g = StateGraph(State)
-    g.add_node("a", node_a)
-    g.add_node("b", node_b)
-    g.add_edge(START, "a")
-    g.add_edge("a", "b")
-    saver = InMemorySaver()
-    graph = g.compile(checkpointer=saver)
-
-    config = {"configurable": {"thread_id": "t1"}}
-    for i in range(6):
-        graph.invoke(
-            {"messages": [HumanMessage(content=f"h{i}", id=f"h{i}")], "other": ""},
-            config,
-        )
-
-    msg_blob_values = [
-        saver.serde.loads_typed((type_tag, blob))
-        for k, (type_tag, blob) in saver.blobs.items()
-        if k[2] == "messages" and type_tag == "msgpack" and blob
-    ]
-    snapshots = [v for v in msg_blob_values if isinstance(v, _DeltaSnapshot)]
-    assert snapshots, "expected at least one _DeltaSnapshot blob for messages"
-
-    state = graph.get_state(config)
-    assert len(state.values["messages"]) == 12  # 6 human + 6 AI
-
-
-def test_delta_channel_snapshot_fires_even_when_not_written() -> None:
-    """Eager snapshot: _DeltaSnapshot stored at snapshot step even when the
-    channel had no write that step (node_b doesn't touch messages).
-    """
-
-    class State(TypedDict):
-        messages: Annotated[
-            list, DeltaChannel(_messages_delta_reducer, snapshot_frequency=3)
-        ]
-        tick: int
-
-    def writer(state: State) -> dict:
-        i = len(state["messages"]) // 2
-        return {"messages": [AIMessage(content=f"a{i}", id=f"a{i}")]}
-
-    def ticker(state: State) -> dict:
-        return {"tick": state["tick"] + 1}
-
-    g = StateGraph(State)
-    g.add_node("writer", writer)
-    g.add_node("ticker", ticker)
-    g.add_edge(START, "writer")
-    g.add_edge("writer", "ticker")
-    saver = InMemorySaver()
-    graph = g.compile(checkpointer=saver)
-
-    config = {"configurable": {"thread_id": "t1"}}
-    for i in range(5):
-        graph.invoke(
-            {"messages": [HumanMessage(content=f"h{i}", id=f"h{i}")], "tick": 0},
-            config,
-        )
-
-    msg_blobs = {
-        k: saver.serde.loads_typed((t, b))
-        for k, (t, b) in saver.blobs.items()
-        if k[2] == "messages" and t == "msgpack" and b
-    }
-    snapshots = {k: v for k, v in msg_blobs.items() if isinstance(v, _DeltaSnapshot)}
-    assert snapshots, (
-        "eager snapshots must fire even on steps where messages wasn't written"
-    )
-
-    state = graph.get_state(config)
-    assert len(state.values["messages"]) == 10  # 5 human + 5 AI
-
-
-# ---------------------------------------------------------------------------
-# DeltaChannel — end-to-end (InMemorySaver)
-# ---------------------------------------------------------------------------
-
-
 def test_delta_channel_inmemory_saver_assembles_writes() -> None:
     """InMemorySaver assembles writes from checkpoint_writes inside get_tuple."""
+    from typing import Annotated
+
+    from langchain_core.messages import AIMessage, HumanMessage
+    from langgraph.checkpoint.memory import InMemorySaver
+    from typing_extensions import TypedDict
+
+    from langgraph.graph import START, StateGraph
+    from langgraph.graph.message import add_messages
 
     class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer, list)]
+        messages: Annotated[list, DeltaChannel(add_messages)]
 
     n = {"v": 0}
 
@@ -369,6 +310,9 @@ def test_delta_channel_inmemory_saver_assembles_writes() -> None:
     graph.invoke({"messages": [HumanMessage(content="hi", id="h1")]}, config)
     graph.invoke({"messages": [HumanMessage(content="bye", id="h2")]}, config)
 
+    # get_tuple returns raw storage shape — channel_values stores DELTA_SENTINEL
+    # for delta channels; the reconstructed writes flow separately via
+    # saver._get_channel_writes_history.
     saved = saver.get_tuple(config)
     assert saved is not None
     assert "messages" in saved.checkpoint["channel_values"]
@@ -378,38 +322,33 @@ def test_delta_channel_inmemory_saver_assembles_writes() -> None:
     assert len(state.values["messages"]) == 4  # 2 human + 2 AI
 
 
-# ---------------------------------------------------------------------------
-# DeltaChannel — dict reducer
-# ---------------------------------------------------------------------------
-
-
-def _delta_channel_with_type(op, typ):
+def _delta_channel_with_type(operator, typ):
     """Build a DeltaChannel with an explicit type via the Annotated injection path."""
-    return _get_channel("_test", Annotated[typ, DeltaChannel(op)])
+    from typing import Annotated
+
+    from langgraph.graph.state import _get_channel
+
+    return _get_channel("_test", Annotated[typ, DeltaChannel(operator)])
 
 
 def test_delta_channel_dict_reducer_fresh_channel() -> None:
     """DeltaChannel with a dict reducer starts as empty dict on MISSING checkpoint."""
 
-    def merge_dicts(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            result.update(w)
-        return result
+    def merge_dicts(left: dict, right: dict) -> dict:
+        return {**left, **right}
 
     ch = _delta_channel_with_type(merge_dicts, dict).from_checkpoint(MISSING)
+    # Should be available (not raise EmptyChannelError) and start empty
     assert ch.is_available()
     assert ch.get() == {}
 
 
 def test_delta_channel_dict_reducer_basic_updates() -> None:
     """DeltaChannel with a dict reducer accumulates key/value pairs across steps."""
+    from langgraph.checkpoint.base import DELTA_SENTINEL
 
-    def merge_dicts(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            result.update(w)
-        return result
+    def merge_dicts(left: dict, right: dict) -> dict:
+        return {**left, **right}
 
     ch = _delta_channel_with_type(merge_dicts, dict).from_checkpoint(MISSING)
 
@@ -427,11 +366,8 @@ def test_delta_channel_dict_reducer_basic_updates() -> None:
 def test_delta_channel_dict_reducer_writes_reconstruction() -> None:
     """replay_writes on a fresh channel replays through a dict merge reducer."""
 
-    def merge_dicts(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            result.update(w)
-        return result
+    def merge_dicts(left: dict, right: dict) -> dict:
+        return {**left, **right}
 
     spec = _delta_channel_with_type(merge_dicts, dict)
     ch = spec.from_checkpoint(DELTA_SENTINEL)
@@ -446,23 +382,29 @@ def test_delta_channel_dict_reducer_writes_reconstruction() -> None:
 
 
 def test_delta_channel_dict_reducer_with_deletions() -> None:
-    """Dict reducer that treats None values as deletions works end-to-end."""
+    """Dict reducer that treats None values as deletions works end-to-end (deepagents pattern)."""
 
-    def merge_files(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            for k, v in w.items():
-                if v is None:
-                    result.pop(k, None)
-                else:
-                    result[k] = v
+    def merge_files(left: dict | None, right: dict) -> dict:
+        if left is None:
+            return {k: v for k, v in right.items() if v is not None}
+        result = {**left}
+        for k, v in right.items():
+            if v is None:
+                result.pop(k, None)
+            else:
+                result[k] = v
         return result
 
     ch = _delta_channel_with_type(merge_files, dict).from_checkpoint(MISSING)
+
     ch.update([{"file1.py": "content1", "file2.py": "content2"}])
+
+    # Delete file1, add file3
     ch.update([{"file1.py": None, "file3.py": "content3"}])
+
     assert ch.get() == {"file2.py": "content2", "file3.py": "content3"}
 
+    # Confirm writes reconstruction produces the same result
     spec = _delta_channel_with_type(merge_files, dict)
     ch2 = spec.from_checkpoint(DELTA_SENTINEL)
     ch2.replay_writes(
@@ -476,27 +418,24 @@ def test_delta_channel_dict_reducer_with_deletions() -> None:
 
 def test_delta_channel_dict_reducer_overwrite_in_update() -> None:
     """Overwrite(dict) in update() must preserve dict shape, not coerce to list."""
+    from langgraph.types import Overwrite
 
-    def merge_dicts(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            result.update(w)
-        return result
+    def merge_dicts(left: dict, right: dict) -> dict:
+        return {**left, **right}
 
     ch = _delta_channel_with_type(merge_dicts, dict).from_checkpoint(MISSING)
     ch.update([{"a": 1}])
     ch.update([Overwrite({"b": 2, "c": 3})])
+
     assert ch.get() == {"b": 2, "c": 3}
 
 
 def test_delta_channel_dict_reducer_overwrite_in_writes_replay() -> None:
     """Overwrite(dict) embedded in replayed writes must reconstruct as dict."""
+    from langgraph.types import Overwrite
 
-    def merge_dicts(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            result.update(w)
-        return result
+    def merge_dicts(left: dict, right: dict) -> dict:
+        return {**left, **right}
 
     spec = _delta_channel_with_type(merge_dicts, dict)
     ch = spec.from_checkpoint(DELTA_SENTINEL)
@@ -511,15 +450,27 @@ def test_delta_channel_dict_reducer_overwrite_in_writes_replay() -> None:
 
 
 def test_delta_channel_dict_reducer_with_notrequired_annotation() -> None:
-    """DeltaChannel infers dict type through `Annotated[NotRequired[dict[...]], ch]`."""
+    """DeltaChannel infers dict type through `Annotated[NotRequired[dict[...]], ch]`.
 
-    def merge_dicts(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            result.update(w)
-        return result
+    This is the shape the deepagents filesystem middleware uses for its
+    `files` field; without unwrapping NotRequired we'd fall through to `list`
+    and blow up on the first dict operator call.
+    """
+    from typing import Annotated
 
-    annotation = Annotated[NotRequired[dict[str, int]], DeltaChannel(merge_dicts)]
+    from typing_extensions import NotRequired
+
+    from langgraph.graph.state import _get_channel
+
+    def merge_dicts(left: dict | None, right: dict) -> dict:
+        if left is None:
+            return dict(right)
+        return {**left, **right}
+
+    annotation = Annotated[
+        NotRequired[dict[str, int]],
+        DeltaChannel(merge_dicts),
+    ]
     ch = _get_channel("files", annotation).from_checkpoint(MISSING)
     assert ch.get() == {}
     ch.update([{"a": 1}])
@@ -528,16 +479,27 @@ def test_delta_channel_dict_reducer_with_notrequired_annotation() -> None:
 
 
 def test_delta_channel_dict_reducer_end_to_end_filesystem() -> None:
-    """End-to-end: graph with dict-reducer (filesystem-style) channel wrapped in DeltaChannel."""
+    """End-to-end: graph with dict-reducer (filesystem-style) channel wrapped in DeltaChannel.
 
-    def merge_files(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            for k, v in w.items():
-                if v is None:
-                    result.pop(k, None)
-                else:
-                    result[k] = v
+    Mirrors the deepagents filesystem pattern: `files: Annotated[dict, reducer]`
+    where the reducer merges dicts and treats None values as deletions.
+    """
+    from typing import Annotated
+
+    from langgraph.checkpoint.memory import InMemorySaver
+    from typing_extensions import TypedDict
+
+    from langgraph.graph import START, StateGraph
+
+    def merge_files(left: dict | None, right: dict) -> dict:
+        if left is None:
+            return {k: v for k, v in right.items() if v is not None}
+        result = {**left}
+        for k, v in right.items():
+            if v is None:
+                result.pop(k, None)
+            else:
+                result[k] = v
         return result
 
     class State(TypedDict):
@@ -560,9 +522,12 @@ def test_delta_channel_dict_reducer_end_to_end_filesystem() -> None:
     for _ in range(3):
         graph.invoke({"files": {}}, config)
 
+    # Checkpoint stores only the sentinel — per-step writes live in checkpoint_writes.
     saved = saver.get_tuple(config)
     assert saved is not None
-    assert saved.checkpoint["channel_values"]["files"] is DELTA_SENTINEL
+    cv = saved.checkpoint["channel_values"]["files"]
+    assert cv is DELTA_SENTINEL
+
     state = graph.get_state(config)
     assert state.values["files"] == {
         "/doc_1.txt": "content for turn 1",
@@ -570,6 +535,7 @@ def test_delta_channel_dict_reducer_end_to_end_filesystem() -> None:
         "/doc_3.txt": "content for turn 3",
     }
 
+    # Deletion path must round-trip through writes replay.
     def delete_file(state: State) -> dict:
         return {"files": {"/doc_1.txt": None}}
 
@@ -590,11 +556,8 @@ def test_delta_channel_dict_reducer_end_to_end_filesystem() -> None:
 def test_delta_channel_dict_reducer_backwards_compat() -> None:
     """A pre-DeltaChannel dict checkpoint must load as a dict, not be listified."""
 
-    def merge_dicts(state: dict, writes: list) -> dict:
-        result = dict(state)
-        for w in writes:
-            result.update(w)
-        return result
+    def merge_dicts(left: dict, right: dict) -> dict:
+        return {**left, **right}
 
     spec = _delta_channel_with_type(merge_dicts, dict)
     old_value = {"a": 1, "b": 2}
@@ -603,7 +566,7 @@ def test_delta_channel_dict_reducer_backwards_compat() -> None:
 
 
 # ---------------------------------------------------------------------------
-# DeltaChannel — seed / pre-delta migration
+# seed / pre-delta migration
 # ---------------------------------------------------------------------------
 
 
@@ -614,7 +577,7 @@ def test_delta_channel_from_checkpoint_honors_seed() -> None:
     a pre-DeltaChannel blob it passes it as `seed` so replay reconstructs
     the post-migration state correctly rather than replaying from empty.
     """
-    spec = DeltaChannel(_messages_delta_reducer, list)
+    spec = DeltaChannel(add_messages)
     seed = [HumanMessage(content="pre-delta", id="p1")]
     ch = spec.from_checkpoint(seed)
     ch.replay_writes(
@@ -630,7 +593,7 @@ def test_delta_channel_from_checkpoint_honors_seed() -> None:
 def test_delta_channel_from_checkpoint_seed_without_writes() -> None:
     """Reconstruction at a pre-delta ancestor with no newer deltas returns
     just the seed — the saver's terminator fired immediately."""
-    spec = DeltaChannel(_messages_delta_reducer, list)
+    spec = DeltaChannel(add_messages)
     seed = [HumanMessage(content="only-snap", id="s1")]
     ch = spec.from_checkpoint(seed)
     ch.replay_writes([])
@@ -644,10 +607,11 @@ def test_delta_channel_from_checkpoint_seed_none_is_distinct_from_sentinel() ->
     explicitly should feed None to the reducer as the left operand.
     """
 
-    def replace(state, writes):
-        return writes[-1] if writes else state
+    def replace(left, right):
+        return right
 
-    spec = DeltaChannel(replace, list)
+    spec = DeltaChannel(replace)
     ch = spec.from_checkpoint(None)
     ch.replay_writes([("t0", "x", "after")])
+    # Reducer replaces; seed=None → first write produces "after".
     assert ch.get() == "after"

libs/langgraph/tests/test_delta_channel_benchmark.py

@@ -1,23 +1,22 @@
-"""Benchmark: DeltaChannel snapshot_frequency — storage vs. read-depth tradeoff.
+"""Benchmark: DeltaChannel vs BinaryOperatorAggregate storage and time.
 
 Run directly:  python tests/test_delta_channel_benchmark.py
 Run via pytest: pytest tests/test_delta_channel_benchmark.py -s
 
-Part 1 — baseline (original): DeltaChannel(inf) vs add_messages (BinOp).
-Part 2 — snapshot_frequency sweep: shows the storage/read-latency tradeoff
-  across frequencies [1, 5, 10, 50, inf] at scale.
+Simulates realistic multi-turn conversations with paragraph-length messages
+(~100 tokens each) scaling up to 1M-token-equivalent histories.
 
-Key insight:
-  snapshot_frequency=inf  → O(N) storage, O(N) read depth (pure delta)
-  snapshot_frequency=N    → O(N²/N) storage, O(N) read depth bounded by freq
-  snapshot_frequency=1    → O(N²) storage, O(1) read depth (full snapshot)
+Token estimates: 1 token ≈ 4 chars; each turn ≈ 200 tokens (human + AI).
+A 1M-token conversation ≈ 5,000 turns of realistic messages.
+
+DeltaChannel stores only a zero-byte sentinel in checkpoint_blobs; the actual
+write data lives in checkpoint_writes (already stored there). Reconstruction
+walks the parent chain and replays writes through the operator — O(N) total
+storage vs O(N²) for plain add_messages.
 """
 
 from __future__ import annotations
 
-import contextlib
-import math
-import os
 import sys
 import time
 from typing import Annotated, Any
@@ -27,17 +26,28 @@ from langchain_core.messages import AIMessage, HumanMessage
 from langgraph.checkpoint.memory import MemorySaver
 from typing_extensions import TypedDict
 
-from langgraph.channels.delta import DeltaChannel
 from langgraph.graph import END, StateGraph
-from langgraph.graph.message import _messages_delta_reducer, add_messages
+from langgraph.graph.message import add_messages
+
+import math as _math_compat
+from langgraph.channels.aggregate import AggregateChannel as _AggregateChannel_compat
+def DeltaChannel(op):
+    return _AggregateChannel_compat(op, snapshot_frequency=_math_compat.inf)
+
+
+try:
+    from langgraph.checkpoint.sqlite import SqliteSaver
+
+    _SQLITE_AVAILABLE = True
+except ImportError:
+    _SQLITE_AVAILABLE = False
 
 try:
     from langgraph.checkpoint.postgres import PostgresSaver
 
     _POSTGRES_AVAILABLE = True
-    _POSTGRES_URI = os.environ.get(
-        "LANGGRAPH_BENCH_POSTGRES_URI",
-        "postgres://postgres@localhost:5432/postgres?sslmode=disable",
+    _POSTGRES_URI = (
+        "postgres://postgres:postgres@localhost:5441/postgres?sslmode=disable"
     )
 except ImportError:
     _POSTGRES_AVAILABLE = False
@@ -118,21 +128,7 @@ class BinaryState(TypedDict):
 
 
 class DeltaState(TypedDict):
-    messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
-
-
-def _make_delta_state(snapshot_frequency: int | float) -> type:
-    """Create a TypedDict with DeltaChannel at the given snapshot_frequency."""
-    channel = DeltaChannel(
-        _messages_delta_reducer, snapshot_frequency=snapshot_frequency
-    )
-    # Use the functional TypedDict form so the Annotated type is stored as an
-    # already-evaluated object rather than a forward-reference string (which
-    # would fail when get_type_hints tries to resolve 'snapshot_frequency').
-    return TypedDict(  # type: ignore[return-value]
-        f"DeltaState_freq{snapshot_frequency}",
-        {"messages": Annotated[list, channel]},
-    )
+    messages: Annotated[list, DeltaChannel(add_messages)]
 
 
 # ---------------------------------------------------------------------------
@@ -178,8 +174,9 @@ def _run_turns(
     """Run n_turns conversation turns.
 
     Returns (write_elapsed_s, read_elapsed_s, total_blob_bytes).
-    Read latency is the average of 5 get_state calls after the full history
-    is built — forces state rehydration including ancestor replay if needed.
+    blob_bytes is -1 for savers without in-memory blob stores (e.g. SQLite).
+    Read latency is measured as the time to invoke the graph with no new
+    messages after the full history is built — this forces state rehydration.
     """
     graph = _make_graph(state_cls, checkpointer)
     config = {"configurable": {"thread_id": "bench"}}
@@ -192,16 +189,16 @@ def _run_turns(
         )
     write_elapsed = time.perf_counter() - t0
 
+    # Measure read/rehydration: get_state forces the channel to rebuild
     t1 = time.perf_counter()
     for _ in range(5):
         graph.get_state(config)
     read_elapsed = (time.perf_counter() - t1) / 5
 
-    blob_bytes = (
-        _total_blob_bytes(graph.checkpointer)
-        if isinstance(graph.checkpointer, MemorySaver)
-        else -1
-    )
+    if isinstance(graph.checkpointer, MemorySaver):
+        blob_bytes = _total_blob_bytes(graph.checkpointer)
+    else:
+        blob_bytes = -1
     return write_elapsed, read_elapsed, blob_bytes
 
 
@@ -214,6 +211,7 @@ def _fmt_bytes(n: int) -> str:
 
 
 def _approx_tokens(n_turns: int) -> str:
+    # ~100 tokens human + ~100 tokens AI per turn
     tokens = n_turns * 200
     if tokens >= 1_000_000:
         return f"~{tokens / 1_000_000:.1f}M tok"
@@ -223,56 +221,74 @@ def _approx_tokens(n_turns: int) -> str:
 
 
 # ---------------------------------------------------------------------------
-# Checkpointer factories
+# Benchmark matrix
 # ---------------------------------------------------------------------------
 
+# Turn counts chosen to demonstrate O(N²) vs O(N) storage growth without running too long.
+# Extrapolation: 5,000 turns × ~200 tokens/turn ≈ 1M tokens (Claude's full context window).
+TURN_COUNTS = [10, 25, 50, 100, 500]
 
-@contextlib.contextmanager
-def _pg_saver(thread_id: str = "bench"):
-    """Context manager that yields a fresh PostgresSaver and cleans up after."""
-    with PostgresSaver.from_conn_string(_POSTGRES_URI) as saver:
-        saver.setup()
-        with saver._cursor() as cur:
-            for tbl in ("checkpoints", "checkpoint_blobs", "checkpoint_writes"):
-                cur.execute(f"DELETE FROM {tbl} WHERE thread_id = %s", (thread_id,))
-        yield saver
-        with saver._cursor() as cur:
-            for tbl in ("checkpoints", "checkpoint_blobs", "checkpoint_writes"):
-                cur.execute(f"DELETE FROM {tbl} WHERE thread_id = %s", (thread_id,))
+# Deep-thread counts where add_messages blob storage would exceed 1 GB;
+# only DeltaChannel runs here.
+DELTA_ONLY_TURN_COUNTS = [1000]
 
 
-def _checkpointers() -> list[tuple[str, Any]]:
-    """Return (label, saver_or_None) pairs for available checkpointers."""
-    result: list[tuple[str, Any]] = [("InMemory", None)]
+def _checkpointer_factories() -> list[tuple[str, Any]]:
+    """Return (label, context_manager_or_none) pairs for available checkpointers."""
+    return [("InMemory", None)]
+
+
+def run_benchmark() -> None:
+    print()
+    print(
+        "DeltaChannel vs add_messages (BinaryOperatorAggregate) — checkpoint storage & latency"
+    )
+    print("Simulating realistic multi-turn conversations up to ~1M-token histories")
+    print("(5,000 turns × ~200 tokens/turn ≈ 1M tokens — Claude's full context window)")
+    print()
+
+    checkpointers: list[tuple[str, Any]] = [("InMemory", None)]
     if _POSTGRES_AVAILABLE:
         try:
             import psycopg
 
             psycopg.connect(_POSTGRES_URI).close()
-            result.append(("Postgres", "postgres"))
+            checkpointers.append(("Postgres (plain SELECT)", "postgres"))
         except Exception:
             pass
-    return result
+
+    for cp_label, cp_hint in checkpointers:
+        print(f"--- Checkpointer: {cp_label} ---")
+        _run_benchmark_for_checkpointer(cp_hint)
 
 
-# ---------------------------------------------------------------------------
-# Part 1: baseline DeltaChannel(inf) vs add_messages
-# ---------------------------------------------------------------------------
-
-BASELINE_TURN_COUNTS = [10, 25, 50, 100, 500]
-DELTA_ONLY_TURN_COUNTS = [1000]
-
-
-def _run_baseline_for_checkpointer(cp_label: str, cp_hint: Any) -> None:
-    W = 72
+def _run_benchmark_for_checkpointer(cp_hint: Any) -> None:
+    import contextlib
+    import tempfile
 
+    @contextlib.contextmanager
     def _make_saver():
         if cp_hint is None:
-            return contextlib.nullcontext(None)
-        return _pg_saver()
+            yield None
+        elif cp_hint == "postgres":
+            with PostgresSaver.from_conn_string(_POSTGRES_URI) as saver:
+                saver.setup()
+                with saver._cursor() as cur:
+                    cur.execute("DELETE FROM checkpoints WHERE thread_id = 'bench'")
+                    cur.execute(
+                        "DELETE FROM checkpoint_blobs WHERE thread_id = 'bench'"
+                    )
+                    cur.execute(
+                        "DELETE FROM checkpoint_writes WHERE thread_id = 'bench'"
+                    )
+                yield saver
+        else:
+            with tempfile.NamedTemporaryFile(suffix=".db") as f:
+                with SqliteSaver.from_conn_string(f.name) as saver:
+                    yield saver
 
     rows: list[tuple[int, Any, Any, Any, Any, Any, Any]] = []
-    for turns in BASELINE_TURN_COUNTS:
+    for turns in TURN_COUNTS:
         with _make_saver() as saver:
             b_wt, b_rt, b_bytes = _run_turns(turns, BinaryState, saver)
         with _make_saver() as saver:
@@ -283,19 +299,26 @@ def _run_baseline_for_checkpointer(cp_label: str, cp_hint: Any) -> None:
             d_wt, d_rt, d_bytes = _run_turns(turns, DeltaState, saver)
         rows.append((turns, None, d_bytes, None, d_rt, None, d_wt))
 
+    # ── Table 1: Storage ─────────────────────────────────────────────────────
+    W = 64
+    print("Storage (checkpoint blob bytes)")
+    print("=" * W)
+    print(
+        f"{'turns':>6}  {'ctx size':>10}  {'add_msgs':>12}  {'delta':>12}  "
+        f"{'savings':>8}"
+    )
+    print("-" * W)
+
     def _bytes_or_na(v: Any) -> str:
-        if v is None or v < 0:
+        if v is None:
+            return "n/a"
+        if v < 0:
             return "n/a"
         return _fmt_bytes(v)
 
     def _ms_or_na(v: Any) -> str:
         return "n/a" if v is None else f"{v * 1000:.1f}ms"
 
-    print(f"\n  [{cp_label}] Storage (blob bytes)")
-    print(
-        f"  {'turns':>6}  {'ctx':>10}  {'add_msgs':>12}  {'delta(inf)':>12}  {'savings':>8}"
-    )
-    print("  " + "-" * (W - 2))
     for turns, b_bytes, d_bytes, b_rt, d_rt, b_wt, d_wt in rows:
         if b_bytes is None or b_bytes < 0 or d_bytes is None or d_bytes < 0:
             ratio_str = "n/a"
@@ -303,132 +326,65 @@ def _run_baseline_for_checkpointer(cp_label: str, cp_hint: Any) -> None:
             ratio = b_bytes / d_bytes if d_bytes else float("inf")
             ratio_str = f"{ratio:.0f}x"
         print(
-            f"  {turns:>6}  {_approx_tokens(turns):>10}  "
-            f"{_bytes_or_na(b_bytes):>12}  {_bytes_or_na(d_bytes):>12}  {ratio_str:>8}"
+            f"{turns:>6}  {_approx_tokens(turns):>10}  "
+            f"{_bytes_or_na(b_bytes):>12}  {_bytes_or_na(d_bytes):>12}  "
+            f"{ratio_str:>8}"
         )
+    print("=" * W)
+    print()
 
-    print(f"\n  [{cp_label}] Read latency (avg of 5 get_state calls)")
-    print(f"  {'turns':>6}  {'ctx':>10}  {'add_msgs':>12}  {'delta(inf)':>12}")
-    print("  " + "-" * (W - 2))
+    # ── Table 2: Read latency ─────────────────────────────────────────────────
+    print("Read latency (avg of 5 get_state calls)")
+    print("=" * W)
+    print(f"{'turns':>6}  {'ctx size':>10}  {'add_msgs':>12}  {'delta':>12}")
+    print("-" * W)
     for turns, b_bytes, d_bytes, b_rt, d_rt, b_wt, d_wt in rows:
         print(
-            f"  {turns:>6}  {_approx_tokens(turns):>10}  "
+            f"{turns:>6}  {_approx_tokens(turns):>10}  "
             f"{_ms_or_na(b_rt):>12}  {_ms_or_na(d_rt):>12}"
         )
-
-
-def run_baseline_benchmark() -> None:
-    print()
-    print("Part 1 — DeltaChannel(inf) vs add_messages: storage & latency")
-    print("=" * 72)
-    for cp_label, cp_hint in _checkpointers():
-        _run_baseline_for_checkpointer(cp_label, cp_hint)
+    print("=" * W)
     print()
 
+    # ── Table 3: Per-invoke latency (total write_elapsed / turns) ─────────────
+    print("Per-invoke latency (total graph.invoke time / turns)")
+    print("=" * W)
+    print(f"{'turns':>6}  {'ctx size':>10}  {'add_msgs':>12}  {'delta':>12}")
+    print("-" * W)
+    for turns, b_bytes, d_bytes, b_rt, d_rt, b_wt, d_wt in rows:
 
-# ---------------------------------------------------------------------------
-# Part 2: snapshot_frequency sweep
-# ---------------------------------------------------------------------------
+        def _per(wt: Any) -> str:
+            if wt is None:
+                return "n/a"
+            return f"{(wt / turns) * 1000:.1f}ms"
 
-# Frequencies to test. 1 = always snapshot (like BinOp), inf = pure delta.
-SNAPSHOT_FREQUENCIES: list[int | float] = [1, 5, 10, 50, math.inf]
-
-# Turn counts for the sweep — high enough to show storage divergence.
-SWEEP_TURN_COUNTS = [50, 100, 500]
-
-
-def _freq_label(freq: int | float) -> str:
-    if freq == math.inf:
-        return "inf"
-    return str(int(freq))
-
-
-def _run_sweep_for_checkpointer(cp_label: str, cp_hint: Any) -> None:
-    def _make_saver():
-        if cp_hint is None:
-            return contextlib.nullcontext(None)
-        return _pg_saver()
-
-    # Collect results: {turns: {freq_label: (write_s, read_s, bytes)}}
-    results: dict[int, dict[str, tuple[float, float, int]]] = {}
-    for turns in SWEEP_TURN_COUNTS:
-        results[turns] = {}
-        for freq in SNAPSHOT_FREQUENCIES:
-            state_cls = _make_delta_state(freq)
-            with _make_saver() as saver:
-                wt, rt, bb = _run_turns(turns, state_cls, saver)
-            results[turns][_freq_label(freq)] = (wt, rt, bb)
-
-    freq_labels = [_freq_label(f) for f in SNAPSHOT_FREQUENCIES]
-    col_w = 12
-
-    header = f"  {'turns':>6}  {'ctx':>10}" + "".join(
-        f"  {f'freq={freq_label}':>{col_w}}" for freq_label in freq_labels
-    )
-
-    print(f"\n  [{cp_label}] Storage (blob bytes) — lower is better")
-    print(header)
-    print("  " + "-" * (len(header) - 2))
-    for turns in SWEEP_TURN_COUNTS:
-        row = f"  {turns:>6}  {_approx_tokens(turns):>10}"
-        for label in freq_labels:
-            _, _, bb = results[turns][label]
-            row += f"  {_fmt_bytes(bb) if bb >= 0 else 'n/a':>{col_w}}"
-        print(row)
-
-    print(f"\n  [{cp_label}] Read latency (avg of 5 get_state) — lower is better")
-    print(header)
-    print("  " + "-" * (len(header) - 2))
-    for turns in SWEEP_TURN_COUNTS:
-        row = f"  {turns:>6}  {_approx_tokens(turns):>10}"
-        for label in freq_labels:
-            _, rt, _ = results[turns][label]
-            row += f"  {f'{rt * 1000:.1f}ms':>{col_w}}"
-        print(row)
-
-    print(
-        f"\n  [{cp_label}] Per-invoke write latency (total / turns) — lower is better"
-    )
-    print(header)
-    print("  " + "-" * (len(header) - 2))
-    for turns in SWEEP_TURN_COUNTS:
-        row = f"  {turns:>6}  {_approx_tokens(turns):>10}"
-        for label in freq_labels:
-            wt, _, _ = results[turns][label]
-            row += f"  {f'{(wt / turns) * 1000:.1f}ms':>{col_w}}"
-        print(row)
-
-
-def run_snapshot_freq_benchmark() -> None:
-    print()
-    print("Part 2 — DeltaChannel snapshot_frequency sweep")
-    print("Lower freq → fewer snapshots → less storage but deeper read replay")
-    print("=" * 80)
-    for cp_label, cp_hint in _checkpointers():
-        _run_sweep_for_checkpointer(cp_label, cp_hint)
+        print(
+            f"{turns:>6}  {_approx_tokens(turns):>10}  "
+            f"{_per(b_wt):>12}  {_per(d_wt):>12}"
+        )
+    print("=" * W)
     print()
+
     print("Legend:")
-    print(
-        "  freq=1    snapshot every write (full blob always — same as add_messages / BinOp)"
-    )
-    print("  freq=N    snapshot every N writes; read walks at most N ancestor writes")
-    print("  freq=inf  pure delta; read walks entire ancestor chain")
+    print("  add_msgs  = Annotated[list, add_messages]  — O(N²) storage")
+    print("  delta     = Annotated[list, DeltaChannel(add_messages)]  — O(N) storage")
     print()
 
 
 # ---------------------------------------------------------------------------
-# Pytest entry points
+# Pytest entry point
 # ---------------------------------------------------------------------------
 
 
 @pytest.mark.skip(
     reason="slow benchmark — run manually with: python tests/test_delta_channel_benchmark.py"
 )
-def test_delta_channel_baseline_benchmark(capsys: Any) -> None:
-    """DeltaChannel(inf) uses less storage than add_messages at scale."""
+def test_delta_channel_benchmark(capsys: Any) -> None:
+    """Storage grows O(N²) for add_messages, O(N) for DeltaChannel."""
     with capsys.disabled():
-        run_baseline_benchmark()
+        run_benchmark()
 
+    # Correctness assertion: DeltaChannel must use less storage at scale.
     for turns in [25, 50]:
         _, _, b_bytes = _run_turns(turns, BinaryState)
         _, _, d_bytes = _run_turns(turns, DeltaState)
@@ -438,41 +394,10 @@ def test_delta_channel_baseline_benchmark(capsys: Any) -> None:
         )
 
 
-@pytest.mark.skip(
-    reason="slow benchmark — run manually with: python tests/test_delta_channel_benchmark.py"
-)
-def test_snapshot_freq_benchmark(capsys: Any) -> None:
-    """snapshot_frequency trades storage for bounded read depth."""
-    with capsys.disabled():
-        run_snapshot_freq_benchmark()
-
-    # Correctness: results at all frequencies should agree on final state.
-    n_turns = 20
-    states: dict[str, list] = {}
-    for freq in SNAPSHOT_FREQUENCIES:
-        state_cls = _make_delta_state(freq)
-        graph = _make_graph(state_cls)
-        config = {"configurable": {"thread_id": "correctness"}}
-        for i in range(n_turns):
-            graph.invoke(
-                {"messages": [HumanMessage(content=_human_content(i), id=f"h{i}")]},
-                config,
-            )
-        state = graph.get_state(config)
-        states[_freq_label(freq)] = [m.id for m in state.values["messages"]]
-
-    ref = states["inf"]
-    for label, msg_ids in states.items():
-        assert msg_ids == ref, (
-            f"freq={label} produced different message IDs than freq=inf"
-        )
-
-
 # ---------------------------------------------------------------------------
 # Script entry point
 # ---------------------------------------------------------------------------
 
 if __name__ == "__main__":
-    run_baseline_benchmark()
-    run_snapshot_freq_benchmark()
+    run_benchmark()
     sys.exit(0)

libs/langgraph/tests/test_delta_channel_migration.py

@@ -46,14 +46,17 @@ import operator
 from typing import Annotated, Any
 
 import pytest
-from langchain_core.messages import AIMessage, HumanMessage
 from langgraph.checkpoint.memory import InMemorySaver
 from typing_extensions import TypedDict
 
 from langgraph.channels.binop import BinaryOperatorAggregate
-from langgraph.channels.delta import DeltaChannel
 from langgraph.graph import END, START, StateGraph
-from langgraph.graph.message import _messages_delta_reducer, add_messages
+
+import math as _math_compat
+from langgraph.channels.aggregate import AggregateChannel as _AggregateChannel_compat
+def DeltaChannel(op):
+    return _AggregateChannel_compat(op, snapshot_frequency=_math_compat.inf)
+
 
 pytestmark = pytest.mark.anyio
 
@@ -72,13 +75,6 @@ def _noop(_state: Any) -> dict:
     return {}
 
 
-def _list_concat(state: list, writes: list) -> list:
-    result = list(state)
-    for w in writes:
-        result.extend(w if isinstance(w, list) else [w])
-    return result
-
-
 def _binop_graph(checkpointer: Any) -> Any:
     class BinopState(TypedDict):
         items: Annotated[list, BinaryOperatorAggregate(list, operator.add)]
@@ -94,7 +90,7 @@ def _binop_graph(checkpointer: Any) -> Any:
 
 def _delta_graph(checkpointer: Any) -> Any:
     class DeltaState(TypedDict):
-        items: Annotated[list, DeltaChannel(_list_concat)]
+        items: Annotated[list, DeltaChannel(operator.add)]
 
     return (
         StateGraph(DeltaState)
@@ -511,103 +507,3 @@ def test_fork_from_update_state_checkpoint() -> None:
         f"fork lost update_state base: base={base_items}, forked={forked_items}"
     )
     assert forked_items[-1] == "fork0", f"fork delta not appended: {forked_items}"
-
-
-# ---------------------------------------------------------------------------
-# 9. Migration from `add_messages` → `DeltaChannel(_messages_delta_reducer)`
-#
-# `add_messages` is the primary real-world use case: it creates a
-# BinaryOperatorAggregate with dedup-by-ID and RemoveMessage semantics.
-# After swapping the annotation to DeltaChannel, pre-migration blobs
-# (plain lists of Message objects) must be used directly as the seed.
-# ---------------------------------------------------------------------------
-
-
-def _add_messages_graph(checkpointer: Any) -> Any:
-    class MessagesState(TypedDict):
-        messages: Annotated[list, add_messages]
-
-    return (
-        StateGraph(MessagesState)
-        .add_node("noop", _noop)
-        .add_edge(START, "noop")
-        .add_edge("noop", END)
-        .compile(checkpointer=checkpointer)
-    )
-
-
-def _delta_messages_graph(checkpointer: Any) -> Any:
-    class DeltaMessagesState(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
-
-    return (
-        StateGraph(DeltaMessagesState)
-        .add_node("noop", _noop)
-        .add_edge(START, "noop")
-        .add_edge("noop", END)
-        .compile(checkpointer=checkpointer)
-    )
-
-
-def test_add_messages_to_delta_migration_preserves_message_history() -> None:
-    """Migration from `add_messages` to `DeltaChannel(_messages_delta_reducer)`
-    preserves message ordering and IDs at both the tip and settled ancestor
-    boundaries.
-
-    The pre-migration blob is a plain list of Message objects; DeltaChannel
-    must use it directly as the seed without walking ancestors past it.
-    """
-    checkpointer = InMemorySaver()
-    config = {"configurable": {"thread_id": "add-messages-migration"}}
-
-    pre_graph = _add_messages_graph(checkpointer)
-    pre_graph.invoke({"messages": [HumanMessage(content="hello", id="h1")]}, config)
-    pre_graph.invoke({"messages": [AIMessage(content="hi", id="a1")]}, config)
-    pre_graph.invoke({"messages": [HumanMessage(content="thanks", id="h2")]}, config)
-
-    pre_tip = pre_graph.get_state(config)
-    assert [m.id for m in pre_tip.values["messages"]] == ["h1", "a1", "h2"]
-
-    delta_graph = _delta_messages_graph(checkpointer)
-
-    # Tip: latest checkpoint has a full list blob — must use it directly.
-    snap = delta_graph.get_state(config)
-    assert [m.id for m in snap.values["messages"]] == ["h1", "a1", "h2"], (
-        f"tip hydration mismatch: got {[m.id for m in snap.values['messages']]}"
-    )
-
-    # Settled ancestor boundaries must also match.
-    pre_settled = [
-        [m.id for m in s.values.get("messages", [])]
-        for s in pre_graph.get_state_history(config)
-        if s.next == ("__start__",)
-    ]
-    delta_settled = [
-        [m.id for m in s.values.get("messages", [])]
-        for s in delta_graph.get_state_history(config)
-        if s.next == ("__start__",)
-    ]
-    assert delta_settled == pre_settled, (
-        f"settled boundary mismatch after migration: "
-        f"pre={pre_settled}, delta={delta_settled}"
-    )
-
-
-async def test_add_messages_to_delta_migration_preserves_message_history_async() -> (
-    None
-):
-    """Async variant of the add_messages migration test."""
-    checkpointer = InMemorySaver()
-    config = {"configurable": {"thread_id": "add-messages-migration-async"}}
-
-    pre_graph = _add_messages_graph(checkpointer)
-    await pre_graph.ainvoke(
-        {"messages": [HumanMessage(content="hello", id="h1")]}, config
-    )
-    await pre_graph.ainvoke({"messages": [AIMessage(content="hi", id="a1")]}, config)
-
-    delta_graph = _delta_messages_graph(checkpointer)
-    snap = await delta_graph.aget_state(config)
-    assert [m.id for m in snap.values["messages"]] == ["h1", "a1"], (
-        f"async tip hydration mismatch: got {[m.id for m in snap.values['messages']]}"
-    )

libs/langgraph/tests/test_pregel.py

@@ -16,7 +16,7 @@ from typing import Annotated, Any, Literal, get_type_hints
 
 import pytest
 from langchain_core.language_models import GenericFakeChatModel
-from langchain_core.messages import AIMessage, AnyMessage, HumanMessage, RemoveMessage
+from langchain_core.messages import AIMessage, AnyMessage, HumanMessage
 from langchain_core.runnables import (
     RunnableConfig,
     RunnableLambda,
@@ -25,7 +25,6 @@ from langchain_core.runnables import (
 from langchain_core.runnables.graph import Edge
 from langgraph.cache.base import BaseCache
 from langgraph.checkpoint.base import (
-    DELTA_SENTINEL,
     BaseCheckpointSaver,
     Checkpoint,
     CheckpointMetadata,
@@ -42,7 +41,6 @@ from typing_extensions import NotRequired, TypedDict
 
 from langgraph._internal._constants import CONFIG_KEY_NODE_FINISHED, ERROR, PULL
 from langgraph.channels.binop import BinaryOperatorAggregate
-from langgraph.channels.delta import DeltaChannel
 from langgraph.channels.ephemeral_value import EphemeralValue
 from langgraph.channels.last_value import LastValue
 from langgraph.channels.topic import Topic
@@ -51,7 +49,7 @@ from langgraph.config import get_stream_writer
 from langgraph.errors import GraphRecursionError, InvalidUpdateError, ParentCommand
 from langgraph.func import entrypoint, task
 from langgraph.graph import END, START, StateGraph
-from langgraph.graph.message import MessagesState, _messages_delta_reducer, add_messages
+from langgraph.graph.message import MessagesState, add_messages
 from langgraph.pregel import (
     NodeBuilder,
     Pregel,
@@ -81,6 +79,14 @@ from tests.messages import (
     _AnyIdToolMessage,
 )
 
+import math as _math_compat
+from langgraph.channels.aggregate import AggregateChannel as _AggregateChannel_compat
+
+
+def DeltaChannel(op):
+    return _AggregateChannel_compat(op, snapshot_frequency=_math_compat.inf)
+
+
 pytestmark = pytest.mark.anyio
 
 logger = logging.getLogger(__name__)
@@ -9406,9 +9412,14 @@ def test_fork_does_not_apply_pending_writes(
 
 async def test_delta_channel_end_to_end_inmemory() -> None:
     """Full graph run: DeltaChannel accumulates correctly across multiple turns."""
+    from langchain_core.messages import AIMessage, HumanMessage
+    from langgraph.checkpoint.memory import InMemorySaver
+
+    from langgraph.graph import START, StateGraph
+    from langgraph.graph.message import add_messages
 
     class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
+        messages: Annotated[list, DeltaChannel(add_messages)]
 
     def respond(state: State) -> dict:
         n = len(state["messages"])
@@ -9442,9 +9453,14 @@ async def test_delta_channel_end_to_end_inmemory() -> None:
 
 async def test_delta_channel_time_travel() -> None:
     """Time-travel back to turn-1 checkpoint and resume; continuation must not include turn-2 deltas."""
+    from langchain_core.messages import AIMessage, HumanMessage
+    from langgraph.checkpoint.memory import InMemorySaver
+
+    from langgraph.graph import START, StateGraph
+    from langgraph.graph.message import add_messages
 
     class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
+        messages: Annotated[list, DeltaChannel(add_messages)]
 
     counter = {"n": 0}
 
@@ -9494,9 +9510,14 @@ async def test_delta_channel_time_travel() -> None:
 
 async def test_delta_channel_remove_message_end_to_end() -> None:
     """RemoveMessage inside a DeltaChannel graph must persist and reload correctly."""
+    from langchain_core.messages import AIMessage, HumanMessage, RemoveMessage
+    from langgraph.checkpoint.memory import InMemorySaver
+
+    from langgraph.graph import START, StateGraph
+    from langgraph.graph.message import add_messages
 
     class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
+        messages: Annotated[list, DeltaChannel(add_messages)]
 
     def respond(state: State) -> dict:
         return {"messages": [AIMessage(content="reply", id="ai-1")]}
@@ -9535,9 +9556,14 @@ async def test_delta_channel_remove_message_end_to_end() -> None:
 
 async def test_delta_channel_update_by_id_end_to_end() -> None:
     """Updating a message by ID via DeltaChannel must persist and reload correctly."""
+    from langchain_core.messages import HumanMessage
+    from langgraph.checkpoint.memory import InMemorySaver
+
+    from langgraph.graph import START, StateGraph
+    from langgraph.graph.message import add_messages
 
     class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
+        messages: Annotated[list, DeltaChannel(add_messages)]
 
     def update_msg(state: State) -> dict:
         # re-send h1 with updated content
@@ -9565,91 +9591,3 @@ async def test_delta_channel_update_by_id_end_to_end() -> None:
     assert "h1" in ids  # h1 persists (updated, not duplicated)
     assert "h2" in ids
     assert ids.count("h1") == 1, "h1 must not be duplicated"
-
-
-async def test_delta_channel_durability_exit_stores_snapshot() -> None:
-    """DeltaChannel must reload from a durability='exit' checkpoint."""
-
-    class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
-
-    def respond(state: State) -> dict:
-        return {"messages": [AIMessage(content="reply", id="ai1")]}
-
-    builder = StateGraph(State)
-    builder.add_node("respond", respond)
-    builder.add_edge(START, "respond")
-    graph = builder.compile(checkpointer=InMemorySaver())
-    config = {"configurable": {"thread_id": "delta-exit-test"}}
-
-    result = graph.invoke(
-        {"messages": [HumanMessage(content="hello", id="h1")]},
-        config,
-        durability="exit",
-    )
-    assert [m.content for m in result["messages"]] == ["hello", "reply"]
-
-    state = graph.get_state(config)
-    assert [m.content for m in state.values["messages"]] == ["hello", "reply"]
-
-
-async def test_delta_channel_async_write_ordering() -> None:
-    """In async mode, DeltaChannel write futures are awaited before the checkpoint
-    is committed, so aput_writes always precedes aput for sentinel checkpoints."""
-
-    class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
-
-    def respond(state: State) -> dict:
-        i = len(state["messages"])
-        return {"messages": [AIMessage(content=f"r{i}", id=f"ai{i}")]}
-
-    order: list[str] = []
-    original_aput_writes = InMemorySaver.aput_writes
-    original_aput = InMemorySaver.aput
-
-    async def tracked_aput_writes(self, config, writes, task_id, task_path=""):
-        result = await original_aput_writes(self, config, writes, task_id, task_path)
-        order.append("aput_writes")
-        return result
-
-    async def tracked_aput(self, config, checkpoint, metadata, new_versions):
-        has_sentinel = any(
-            v is DELTA_SENTINEL for v in checkpoint.get("channel_values", {}).values()
-        )
-        order.append("aput_sentinel" if has_sentinel else "aput_other")
-        return await original_aput(self, config, checkpoint, metadata, new_versions)
-
-    InMemorySaver.aput_writes = tracked_aput_writes
-    InMemorySaver.aput = tracked_aput
-    try:
-        builder = StateGraph(State)
-        builder.add_node("respond", respond)
-        builder.add_edge(START, "respond")
-        graph = builder.compile(checkpointer=InMemorySaver())
-        config = {"configurable": {"thread_id": "async-ordering-test"}}
-
-        for i in range(3):
-            await graph.ainvoke(
-                {"messages": [HumanMessage(content=f"h{i}", id=f"h{i}")]}, config
-            )
-
-        # Every aput_sentinel must be preceded by at least one aput_writes
-        for i, event in enumerate(order):
-            if event == "aput_sentinel":
-                preceding = order[:i]
-                assert "aput_writes" in preceding, (
-                    f"aput_sentinel at {i} had no preceding aput_writes: {order}"
-                )
-                last_write_idx = max(
-                    j for j, e in enumerate(order[:i]) if e == "aput_writes"
-                )
-                assert last_write_idx < i, (
-                    f"aput_writes at {last_write_idx} should precede aput_sentinel at {i}: {order}"
-                )
-    finally:
-        InMemorySaver.aput_writes = original_aput_writes
-        InMemorySaver.aput = original_aput
-
-    state = await graph.aget_state(config)
-    assert len(state.values["messages"]) == 6  # 3 human + 3 AI

libs/langgraph/tests/test_pregel_async.py

@@ -6101,36 +6101,6 @@ async def test_parent_command(
     )
 
 
-async def test_delta_channel_durability_exit_stores_snapshot_async() -> None:
-    """DeltaChannel must reload from an async durability='exit' checkpoint."""
-    from langchain_core.messages import AIMessage
-
-    from langgraph.channels.delta import DeltaChannel
-    from langgraph.graph.message import _messages_delta_reducer
-
-    class State(TypedDict):
-        messages: Annotated[list, DeltaChannel(_messages_delta_reducer)]
-
-    async def respond(state: State) -> dict:
-        return {"messages": [AIMessage(content="reply", id="ai1")]}
-
-    builder = StateGraph(State)
-    builder.add_node("respond", respond)
-    builder.add_edge(START, "respond")
-    graph = builder.compile(checkpointer=InMemorySaver())
-    config = {"configurable": {"thread_id": "delta-exit-async-test"}}
-
-    result = await graph.ainvoke(
-        {"messages": [HumanMessage(content="hello", id="h1")]},
-        config,
-        durability="exit",
-    )
-    assert [m.content for m in result["messages"]] == ["hello", "reply"]
-
-    state = await graph.aget_state(config)
-    assert [m.content for m in state.values["messages"]] == ["hello", "reply"]
-
-
 @NEEDS_CONTEXTVARS
 async def test_interrupt_subgraph(async_checkpointer: BaseCheckpointSaver) -> None:
     class State(TypedDict):

libs/langgraph/uv.lock

@@ -7,9 +7,6 @@ resolution-markers = [
     "python_full_version < '3.11'",
 ]
 
-[options]
-prerelease-mode = "allow"
-
 [[package]]
 name = "aiosqlite"
 version = "0.22.1"
@@ -1351,11 +1348,10 @@ wheels = [
 
 [[package]]
 name = "langchain-core"
-version = "1.3.2"
+version = "1.3.0"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "jsonpatch" },
-    { name = "langchain-protocol" },
     { name = "langsmith" },
     { name = "packaging" },
     { name = "pydantic" },
@@ -1364,26 +1360,14 @@ dependencies = [
     { name = "typing-extensions" },
     { name = "uuid-utils" },
 ]
-sdist = { url = "https://files.pythonhosted.org/packages/a8/03/7219502e8ca728d65eb44d7a3eb60239230742a70dbfc9241b9bfd61c4ab/langchain_core-1.3.2.tar.gz", hash = "sha256:fd7a50b2f28ba561fd9d7f5d2760bc9e06cf00cdf820a3ccafe88a94ffa8d5b7", size = 911813, upload-time = "2026-04-24T15:49:23.699Z" }
+sdist = { url = "https://files.pythonhosted.org/packages/92/fe/20190232d9b513242899dbb0c2bb77e31b4d61e343743adbe90ebc2603d2/langchain_core-1.3.0.tar.gz", hash = "sha256:14a39f528bf459aa3aa40d0a7f7f1bae7520d435ef991ae14a4ceb74d8c49046", size = 860755, upload-time = "2026-04-17T14:51:38.298Z" }
 wheels = [
-    { url = "https://files.pythonhosted.org/packages/7d/d5/8fa4431007cbb7cfed7590f4d6a5dea3ad724f4174d248f6642ef5ce7d05/langchain_core-1.3.2-py3-none-any.whl", hash = "sha256:d44a66127f9f8db735bdfd0ab9661bccb47a97113cfd3f2d89c74864422b7274", size = 542390, upload-time = "2026-04-24T15:49:21.991Z" },
-]
-
-[[package]]
-name = "langchain-protocol"
-version = "0.0.14"
-source = { registry = "https://pypi.org/simple" }
-dependencies = [
-    { name = "typing-extensions" },
-]
-sdist = { url = "https://files.pythonhosted.org/packages/05/bf/efb5e2ed832e4d6d45590e25a9e5191986b291b543bc6a807b48bee070b0/langchain_protocol-0.0.14.tar.gz", hash = "sha256:bc1e8553122e6ede310280462d5813023a172ff2785ccbbdec54d43f3a15e5f2", size = 5862, upload-time = "2026-04-29T16:40:18.657Z" }
-wheels = [
-    { url = "https://files.pythonhosted.org/packages/c2/e9/06c47ecb2aff08f83dfa30058da3bf86be64862c19569043ed5331bbeecd/langchain_protocol-0.0.14-py3-none-any.whl", hash = "sha256:ffc35089779bd8ca217015180cef5e660fc3b074efdaa0f2e95df73583f1a047", size = 6984, upload-time = "2026-04-29T16:40:17.841Z" },
+    { url = "https://files.pythonhosted.org/packages/f8/e2/dbfa347aa072a6dc4cd38d6f9ebfc730b4c14c258c47f480f4c5c546f177/langchain_core-1.3.0-py3-none-any.whl", hash = "sha256:baf16ee028475df177b9ab8869a751c79406d64a6f12125b93802991b566cced", size = 515140, upload-time = "2026-04-17T14:51:36.274Z" },
 ]
 
 [[package]]
 name = "langgraph"
-version = "1.2.0a2"
+version = "1.1.9"
 source = { editable = "." }
 dependencies = [
     { name = "langchain-core" },
@@ -1455,7 +1439,7 @@ test = [
 
 [package.metadata]
 requires-dist = [
-    { name = "langchain-core", specifier = ">=1.3.2,<2" },
+    { name = "langchain-core", specifier = ">=1.3.0,<2" },
     { name = "langgraph-checkpoint", editable = "../checkpoint" },
     { name = "langgraph-prebuilt", editable = "../prebuilt" },
     { name = "langgraph-sdk", editable = "../sdk-py" },
@@ -1564,7 +1548,7 @@ wheels = [
 
 [[package]]
 name = "langgraph-checkpoint"
-version = "4.1.0a2"
+version = "4.0.2"
 source = { editable = "../checkpoint" }
 dependencies = [
     { name = "langchain-core" },

notes/delta_channel_query_bench.md

@@ -0,0 +1,137 @@
+# Delta-channel reconstruction: query strategy benchmark
+
+**Branch:** `delta-channel-writes-based`
+**Question (Nuno):** Is the recursive CTE the right query shape for reconstructing a delta channel inside `get_tuple`, or would a plain `SELECT WHERE` be cheaper even though it returns more rows?
+**Answer:** Plain `SELECT WHERE` wins at every realistic depth. The recursion isn't the problem — the JSON-expression join inside the CTE is.
+
+## Setup
+
+- Postgres 16 on `localhost:5441` (the `compose-postgres.yml` instance, run directly without docker for this round)
+- Single delta channel `messages`, one write per checkpoint, `DELTA_SENTINEL` blob per checkpoint
+- Linear chain (`branch=1`) and 5-way branching at every step (`branch=5`) — branching is the case where plain over-fetches sibling rows
+- Median of 20 timed runs after 3 warmups, fresh psycopg cursor per strategy
+- Bench script: `bench_get_tuple_strategies.py` at repo root
+
+Three strategies compared:
+
+| name | roundtrips | shape |
+|------|-----------|-------|
+| `cte` | 1 | Current prod: recursive CTE walks ancestors, LEFT JOINs writes + blobs |
+| `plain` | 3 | Nuno's suggestion: thread-wide `SELECT WHERE` per table, Python walks parent chain and filters |
+| `cte+narrow` | 2 | CTE returns ancestor IDs only, then one `UNION ALL` of writes + blobs filtered by `ANY(ids)` |
+
+## Results (ms per get_tuple, median of 20)
+
+```
+ depth  branch       cte       plain    cte+narrow   rows_cte   rows_plain   plain/cte
+    10       1     0.14ms      0.26ms       0.24ms          9           30      1.89x
+    10       5     0.21ms      0.23ms       0.17ms          9          110      1.11x
+    50       1     0.89ms      0.27ms       0.33ms         49          150      0.31x
+    50       5     2.35ms      0.66ms       0.51ms         49          550      0.28x
+   200       1    11.61ms      0.78ms       1.30ms        199          600      0.07x
+   200       5    34.79ms      2.33ms       3.07ms        199         2200      0.07x
+  1000       1   274.60ms      2.59ms      13.29ms        999         3000      0.01x
+  1000       5   856.01ms     10.14ms      15.31ms        999        11000      0.01x
+```
+
+Lower is better. `plain/cte < 1` means plain is faster.
+
+### Headline numbers
+
+- depth 50: plain is **3x** faster
+- depth 200: plain is **15x** faster
+- depth 1000: plain is **~100x** faster
+- Branching makes plain over-fetch (3000 rows → 11000 rows at d=1000), but it remains ~85x faster than the CTE
+
+## Why the CTE collapses
+
+`EXPLAIN (ANALYZE, BUFFERS)` of the CTE at depth 1000 (linear). Excerpt with the load-bearing nodes:
+
+```
+Sort  ... actual time=137.798..137.827 rows=999
+  CTE ancestors
+    ->  Recursive Union  ... actual time=0.005..2.443 rows=999
+                                                 ^^^^^^
+                                                 recursion is 2.4 ms — fine
+  ->  Nested Loop Left Join  ... actual time=2.676..137.529 rows=999
+        Join Filter: (cw.checkpoint_id = a.cid)
+        Rows Removed by Join Filter: 998001
+                                     ^^^^^^^
+                                     999 ancestors x ~1000 writes
+        ->  Nested Loop Left Join  ... actual time=2.669..85.061 rows=999
+              Join Filter: (bl.version = ((c.checkpoint -> 'channel_versions'::text) ->> bl.channel))
+              Rows Removed by Join Filter: 998001
+                                           ^^^^^^^
+                                           same quadratic blow-up on the blob join
+```
+
+Two pathological things are happening:
+
+1. **The blob join filter is on a JSON expression**: `bl.version = (c.checkpoint -> 'channel_versions' ->> bl.channel)`. The planner cannot push this into an index lookup, so it materializes `checkpoint_blobs` for the thread and does a nested-loop comparison against every ancestor — a Cartesian product that grows as `O(ancestors × blobs_in_thread)`.
+2. **The writes join is similar**: writes for the thread are materialized once, then nested-loop joined against ancestors with a `Join Filter` rather than a hash/merge join over the indexed `checkpoint_id`.
+
+At depth 1000 that's **~2 million rows evaluated, 99.9% of them discarded**. The recursion itself is a rounding error.
+
+For comparison, the plain Q1 (`SELECT … FROM checkpoints WHERE thread_id=? AND checkpoint_ns=?`) at depth 1000:
+
+```
+Seq Scan on checkpoints  ... actual time=0.012..0.121 rows=1000
+Execution Time: 0.140 ms
+```
+
+A simple seq scan over 57 buffers. Q2 and Q3 follow the same shape and complete in well under 1 ms each.
+
+## Crossover and remote-DB reasoning
+
+- Pure local Postgres: plain wins from depth ~30 onward; CTE wins by fractions of a ms below that
+- Remote Postgres at ~5 ms RTT adds ~10 ms to plain (3 roundtrips vs 1). Crossover shifts to ~depth 30. Above that, the CTE's quadratic SQL cost still dominates the RTT savings.
+
+There is no realistic conversation depth where the CTE wins on a remote DB. At depth 200+ (anything resembling a real multi-turn agent run) plain is faster regardless of network.
+
+## Recommendation
+
+**Switch to plain SELECT WHERE, one delta channel at a time.**
+
+Three indexed queries per delta channel:
+
+```sql
+-- Q1: parent chain + per-checkpoint version of this channel
+SELECT checkpoint_id,
+       parent_checkpoint_id,
+       checkpoint -> 'channel_versions' ->> 'channel_name' AS ver
+FROM checkpoints
+WHERE thread_id = ? AND checkpoint_ns = ?;
+
+-- Q2: writes for this channel, anywhere in the thread
+SELECT checkpoint_id, type, blob, task_id, idx
+FROM checkpoint_writes
+WHERE thread_id = ? AND checkpoint_ns = ? AND channel = ?;
+
+-- Q3: blobs for this channel, anywhere in the thread
+SELECT version, type, blob
+FROM checkpoint_blobs
+WHERE thread_id = ? AND checkpoint_ns = ? AND channel = ?;
+```
+
+Python then:
+- Builds `parent_of: dict[cid, parent_cid]` from Q1
+- Walks from target's parent newest → oldest
+- Filters Q2 rows by `ancestor_set`, processes oldest → newest, applies overwrite-terminator
+- Picks seed blob via the per-ancestor `ver` map, terminates at first non-sentinel blob
+
+All O(n) on n = thread checkpoints, with tight constants (dict lookups). No recursion, no JSON-expression joins, no quadratic plans.
+
+If the 3-roundtrip cost ever shows up on remote-DB benchmarks, fold Q2 + Q3 into one `UNION ALL` to get back to 2 roundtrips. Bench says it isn't worth the SQL complexity right now.
+
+## Bonus: code simplification from single-channel scope
+
+Multi-channel reconstruction in the current `_reconstruct_delta_channels_cur` carries:
+
+- `rows_by_cid` nested dicts, keyed by cid then channel
+- `seen_blob: set[(cid, channel)]` and `seen_write: set[(cid, channel, task_id, idx)]` dedup
+- `collected: dict[channel, list]`, `done: set[channel]`, `seeds: dict[channel, value]`
+- Inner `for ch in channels_list` loops and an early-exit `if len(done) == len(channels_list)`
+
+Single-channel collapses these to a single list, a single bool, and one `Optional[Any]`. Roughly half the Python in that function, plus an obvious shape for splitting pure post-processing into `base.py` so sync and async stop duplicating it.
+
+If multi-channel coalescing turns out to matter later, it can come back as a SQL-level optimization without re-introducing the bookkeeping in Python.