Nick Jamesandpc.yu c82ae5f664 fix(embedding): harden request-scoped query embedding cache (#4822) ...

Follow-up to #4794: install the cache scope at the fan-out owners via a
query_embed_cache_scope context manager (covers /recall and the MCP
search tool), cover the /skills/find two-find fan-out, shield the shared
embed task against waiter cancellation, key entries by embedder identity,
and reset the ContextVar token on scope exit.

Co-authored-by: pc.yu <nick@fourieralpha.com>

2026-09-09 19:21:10 +08:00

..

vlm

feat: 支持基于火山方舟的音视频多模态理解 (#3563)

2026-08-03 16:44:22 +08:00

test_embedding_telemetry_usage.py

fix(volcengine): update default doubao embedding model (#1438)

2026-04-14 15:51:06 +08:00

test_openai_rerank.py

fix(rerank): send top_n in OpenAI-compatible rerank requests (#4040)

2026-08-26 11:19:21 +08:00

test_query_embedding_cache.py

fix(embedding): harden request-scoped query embedding cache (#4822)

2026-09-09 19:21:10 +08:00

test_vlm_strip_think_tags.py

refactor(model): unify config-driven retry across VLM and embedding (#926)

2026-03-31 16:17:28 +08:00