wangweiandCopilot f6e10145cd Fix: build_models resolves separate AsyncOpenAI clients for judge and embedding models ...

Previously, only the judge_model's profile was looked up and its AsyncOpenAI

client was shared with embedding_factory. When embedding_model has a different

base_url/api_key (e.g. Qwen3-Embedding-4B on SiliconFlow vs gpt-5 on another

gateway), the embedding calls silently used the wrong URL, causing 402/404 errors.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

2026-07-01 21:15:09 +08:00

..

__init__.py

first commit

2026-06-12 14:02:15 +08:00

factory.py

Fix: build_models resolves separate AsyncOpenAI clients for judge and embedding models

2026-07-01 21:15:09 +08:00

judge_prompts.py

Add committed zh judge-prompt cache + enable in Siemens scenarios + docs

2026-07-01 18:15:53 +08:00

pipeline.py

更新

2026-06-16 18:12:33 +08:00

registry.py

更新

2026-06-16 18:12:33 +08:00

weights.py

feat: add metric/doc weight computation module (weights.py)

2026-06-18 16:47:47 +08:00