WEBDIENSTE/openswarm
1
0
Fork 0
mirror of https://github.com/openswarm-ai/openswarm.git synced 2026-08-30 11:49:50 +02:00
Code Issues Packages Projects Releases Wiki Activity

[eric] distill: use the reliable cheap aux tier, not the chat's family (gemini/codex sessions 404'd/401'd the same-family aux)

Browse Source
This commit is contained in:
ciregenz
2026-07-06 18:09:34 -07:00
parent fd21ffb603
commit 95f1996c19
1 changed files with 3 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/apps/agents/manager/session/distill_history.py
+3 -4
View File
@@ -93,12 +93,11 @@ async def distilled_history_summary(session: AgentSession, settings: AppSettings
@typechecked
async def p_call_distiller(session: AgentSession, settings: AppSettings, body: str) -> str:
from backend.apps.agents.providers.registry import resolve_aux_model, get_api_type
from backend.apps.agents.providers.registry import resolve_aux_model
from backend.apps.settings.credentials import get_anthropic_client_for_model
aux_model, _ = await resolve_aux_model(
settings, preferred_tier="haiku", primary_api=get_api_type(session.model),
)
# No primary_api: a background summary wants the most RELIABLE cheap tier, not the chat's family. Forcing the family routed a gemini/codex chat's distill onto a same-family aux that 404s (gemini-direct google endpoint the Anthropic client can't call) or 401s (codex token rotation); the proven classifier omits it too and resolves to whatever anthropic-compatible lane the user has.
aux_model, _ = await resolve_aux_model(settings, preferred_tier="haiku")
client = get_anthropic_client_for_model(settings, aux_model)
resp = await client.messages.create(
model=aux_model,