WEBDIENSTE/langgraph
1
0
Fork 0
mirror of https://github.com/langchain-ai/langgraph.git synced 2026-08-29 19:29:43 +02:00
Code Issues Packages Projects Releases Wiki Activity
4,022 Commits 322 Branches 572 Tags
9fe61a42d2bf63c4ba4a94de71e6e8016729a951
Commit Graph
1642 Commits
Author SHA1 Message Date
Nuno CamposandGitHub 8a80b1d3b1 Add 2 more benchmark graphs (#1707) 2024-09-12 16:31:16 -07:00
Nuno CamposandGitHub 95b97d69ec Nc/12sep/perf (#1702) 
* Remove unused fil;e

* Add benchmark-fast command for running locally

* Small improvements to jsonplus serializer

* Don't use PregelNode.mapper when schema is a typed dict

- All it would do is create a new copy of same dict

* Avoid copying checkpoint when fetching at beginning of loop

* Fix needs array

* Update tests
 2024-09-12 12:50:53 -07:00
Nuno CamposandGitHub 8f87cd1c91 Performance improvements in core library (#1683) 
* Performance improvements in core library

- Avoid creating new callback manager when received one as arg
- Avoid looking for config when already received one as arg
- Avoid copies of values in ensure_config/merge_configs
- Implement version of ensure_config that accepts multiple configs (avoids calling merge_configs first)
- Avoid calling merge_configs when we only need to attach extra tags/metadata

* Fix

* Fix

* Try again

* Debug ci job

* Fix

* Try again

* Try again

* Try again

* Some more variations

* Attach annotation to first changed file

* Fix

* Re-enable benchmarks
 2024-09-12 11:01:37 -07:00
Nuno CamposandGitHub 73d1051034 Merge pull request #1693 from langchain-ai/nc/11sep/kafka-finally-send 
kafka: Expose finally_send as public api
 2024-09-11 16:48:41 -07:00
Nuno Campos 571623442a Avoid closing drain before seeing first message in either topic 2024-09-11 16:45:43 -07:00
Nuno CamposandGitHub c172362918 Merge pull request #1691 from langchain-ai/nc/11sep/kafka-sync 
kafka: Implement sync orchestrator and executor classes
 2024-09-11 16:34:01 -07:00
Nuno CamposandGitHub 115c37acea Merge pull request #1686 from langchain-ai/nc/11sep/kafka-configurable-consumer-producer 
kafka: Make consumer and producer classes configurable
 2024-09-11 16:33:15 -07:00
Nuno Campos dd68babac7 kafka: Expose finally_send as public api 
- Enables attaching a list of messages (topic/key/value) to send when run finishes
 2024-09-11 16:31:35 -07:00
Vadym BardaandGitHub 63c34ca992 langgraph: update snapshots for 0.3.x core tests (#1692) 2024-09-11 19:28:47 -04:00
Nuno Campos 4accc19734 kafka: Implement sync orchestrator and executor classes 2024-09-11 16:06:50 -07:00
Vadym BardaandGitHub 84643b2789 langgraph: update core & snapshots (#1689) 2024-09-11 21:59:12 +00:00
Nuno Campos 217b07ae65 kafka: Make consumer and producer classes configurable 
- Define protocol for sync and async producer and consumer
- Accept consumer/producer as init args in Orchestrator/Executor
- If not passed in, create default consumer/producer as before
 2024-09-11 14:27:08 -07:00
Nuno Campos f179da00a0 update bench job 2024-09-11 13:41:45 -07:00
Nuno CamposandGitHub 8e275fa485 Merge pull request #1609 from langchain-ai/nc/4sep/benchmark 
Add benchmark script
 2024-09-11 12:29:22 -07:00
Nuno Campos 47dae71841 Remove files 2024-09-11 12:28:38 -07:00
Nuno Campos bf4f8e67d8 Rename command 2024-09-11 12:25:22 -07:00
Nuno Campos d844359d61 Fix output dir 2024-09-11 12:23:59 -07:00
Nuno Campos f577529356 Add benchmark/profile commands 2024-09-11 11:44:18 -07:00
Tat Dat Duong a60a4606fb fix(sdk-js): bump to 0.0.9 2024-09-11 19:41:51 +02:00
gbaian10andGitHub bf6ad19963 refactor: Type narrowing for ToolNode.name (#1678) 
Avoid type incompatibility when using `add_node` to add a `ToolNode`.
 2024-09-11 13:11:39 -04:00
Vadym BardaandGitHub f3b5d804ab ci: fix kafka scheduler, split langgraph and kafka into separate test jobs (#1677) 2024-09-11 10:55:11 -04:00
Nuno Campos 26f23ea26c Add benchmark script 2024-09-10 17:32:47 -07:00
Nuno Campos 5de46f7472 Update stop condition for drain 2024-09-10 16:52:09 -07:00
Nuno Campos bdabb723b0 Add langgraph dep 2024-09-10 16:39:34 -07:00
Nuno Campos edc0fd0210 Remove extraneous asserts 2024-09-10 16:20:45 -07:00
Nuno Campos 3886b76771 Lint 2024-09-10 16:17:22 -07:00
Nuno Campos 3a65460767 Use anyio for timeout in dran test helper 2024-09-10 16:17:22 -07:00
Nuno Campos 5e0273886e Lint 2024-09-10 16:17:22 -07:00
Nuno Campos 943cc4d83a Add readme, lint 2024-09-10 16:17:22 -07:00
Nuno Campos 270559b880 Implement changes in sqlite checkpointer 2024-09-10 16:17:22 -07:00
Nuno Campos 98b8595a94 Move to dev deps 2024-09-10 16:17:22 -07:00
Nuno Campos f7dffa023c Implement subgraph delegation for distributed arch 2024-09-10 16:17:22 -07:00
Nuno Campos 7dc69a543d Rename 2024-09-10 16:17:22 -07:00
Nuno Campos 8f91402642 Use thread_id for partition key, ignore tasks for stale checkpoints 2024-09-10 16:17:22 -07:00
Nuno Campos 030d6d2def Fix subgraph test 2024-09-10 16:17:22 -07:00
Nuno Campos ad7dd0dbb7 Add drain_topics test helper 2024-09-10 16:17:22 -07:00
Nuno Campos 0eb11a3cf9 Remove is_resuming arg 2024-09-10 16:17:22 -07:00
Nuno Campos 1ce8d628f2 In update state skip nodes not found in instance 
eg, because removed since
 2024-09-10 16:17:22 -07:00
Nuno Campos 5e6b58f847 Remove double shield usage 
- we already shield from cancellation in PregelLoop so shouldnt be repeating in AsyncBackgroundExecutor
 2024-09-10 16:17:22 -07:00
Nuno Campos b13db42568 WIP Subgraphs 2024-09-10 16:17:22 -07:00
Nuno Campos 83d36761a4 Rename compose projects 2024-09-10 16:17:22 -07:00
Nuno Campos d206621230 Rename test 2024-09-10 16:17:22 -07:00
Nuno Campos 8e6caab78c Add test for interrupts 2024-09-10 16:17:22 -07:00
Nuno Campos ad463ec90a Add more test assertions 2024-09-10 16:17:22 -07:00
Nuno Campos 2d02e5e0c5 Add error topic, add retries (eg for checkpointer database exceptions) 2024-09-10 16:17:22 -07:00
Nuno Campos 5b3bd920d8 Catch errors thrown by nodes run by Executor 2024-09-10 16:17:22 -07:00
Nuno Campos f09b130837 Use pool for postgres checkpointer 2024-09-10 16:17:22 -07:00
Nuno Campos e6d6f565a5 Improvements to consumer and producer patterns 
- await future returned by send() instead of flush()
- use consumer groups by default
- process tasks in batches by default, configurable
- manually commit offsets when batch is processed
 2024-09-10 16:17:22 -07:00
Nuno Campos dc5be3fd58 Rename 2024-09-10 16:17:22 -07:00
Nuno Campos 381ea9cec8 Rename 2024-09-10 16:17:22 -07:00