Kevin 80833f7033 feat(api): DeepSeek V4 Flash 默认、HTTP 错讯与多供应商分层 ...

- 主链路默认 deepseek-v4-flash，DEEPSEEK_THINKING_ENABLED 对齐旧非思考 chat
- 评测台评审装配迁入 adapters/llm（deepseek_eval_judge、zhipu_eval_judge）与 eval_judge_spec
- 拆分 llm_http_openai_chat_errors 与 llm_errors（DeepSeek/智谱品牌与文档链），llm_call 支持 http_error_vendor
- EvalJudgeService 按 spec.provider 传入 allm_json_call；评测台前端文案改为 V4 Flash
- 更新 .env 示例与 staging/production 的 DEEPSEEK_MODEL；补充 openai/供应商错讯测试

Made-with: Cursor

2026-04-27 14:34:30 +08:00

..

importers

feat/ eval

2026-04-06 23:19:20 +08:00

rubrics

WIP: memory system improvements (in progress)

2026-04-22 16:56:28 +08:00

__init__.py

feat/ 导出开发容器内的数据用于评估

2026-04-03 14:44:46 +08:00

admin_service.py

refactor(eval+memoir)：精简内部评测路由与服务，composite/对话摘要与 judge 能力补强

2026-04-08 21:36:12 +08:00

composite_score.py

refactor(eval+memoir)：精简内部评测路由与服务，composite/对话摘要与 judge 能力补强

2026-04-08 21:36:12 +08:00

conversation_compare_summary.py

feat:

2026-04-09 15:32:35 +08:00

deps.py

feat(evaluation): memoir readiness, judge/replay updates, eval web playground

2026-04-08 09:43:34 +08:00

errors.py

feat/ 导出开发容器内的数据用于评估

2026-04-03 14:44:46 +08:00

eval_trace_format.py

Merge branch 'eval/elapsed-time-memoir-batch-chunk' into development

2026-04-10 10:27:41 +08:00

eval_trace_repo.py

feat:

2026-04-09 15:32:35 +08:00

eval_trace_schemas.py

feat: 回忆录证据血缘与内部评测可追溯，顺带对齐本地评测台与 CI

2026-04-08 15:37:09 +08:00

eval_trace_service.py

feat:

2026-04-09 15:32:35 +08:00

gate_report_service.py

feat/ 导出开发容器内的数据用于评估

2026-04-03 14:44:46 +08:00

gating_service.py

feat/ 导出开发容器内的数据用于评估

2026-04-03 14:44:46 +08:00

internal_auth.py

feat/ 导出开发容器内的数据用于评估

2026-04-03 14:44:46 +08:00

judge_manual_service.py

feat(api): DeepSeek V4 Flash 默认、HTTP 错讯与多供应商分层

2026-04-27 14:34:30 +08:00

judge_schemas.py

fix:

2026-04-10 20:35:57 +08:00

judge_service.py

feat(api): DeepSeek V4 Flash 默认、HTTP 错讯与多供应商分层

2026-04-27 14:34:30 +08:00

memoir_compare_summary.py

fix:

2026-04-10 20:35:57 +08:00

memoir_readiness_service.py

Merge branch 'eval/elapsed-time-memoir-batch-chunk' into development

2026-04-10 10:27:41 +08:00

models.py

feat/ 导出开发容器内的数据用于评估

2026-04-03 14:44:46 +08:00

phase1_job_timing.py

feat(eval): server-side replay/phase1 timing + memoir phase1 batch chunking

2026-04-09 13:39:04 +08:00

replay_service.py

Merge branch 'eval/elapsed-time-memoir-batch-chunk' into development

2026-04-10 10:27:41 +08:00

repo.py

feat: 回忆录证据血缘与内部评测可追溯，顺带对齐本地评测台与 CI

2026-04-08 15:37:09 +08:00

router.py

Merge branch 'eval/elapsed-time-memoir-batch-chunk' into development

2026-04-10 10:27:41 +08:00

schemas.py

feat(eval): memoir A/B chapter judging and eval-web parity with dialogue

2026-04-10 10:25:15 +08:00

session_catalog_repo.py

feat(evaluation): session catalog, user export import, and eval web UI

2026-04-06 13:49:28 +08:00

session_catalog_service.py

feat(eval): Playground GLM 评分落库并可恢复

2026-04-08 16:51:08 +08:00

transcript_for_judge.py

feat:

2026-04-09 15:32:35 +08:00

user_export_fixtures.py

feat(evaluation): session catalog, user export import, and eval web UI

2026-04-06 13:49:28 +08:00