yumoqing 7e4069f3b6 refactor: get_llm() uses get_llmage_llm() + cached uapi/uapiio lookups ...

Replace 6-table JOIN with 3-step approach:
1. get_llmage_llm() for base info (llm + llm_api_map + llmcatelog)
2. Cached uapi lookup (ioid, stream, callbackurl)
3. Cached uapiio lookup (input_fields)

Benefits:
- Code reuse: eliminates duplicate SQL
- Performance: uapi/uapiio cached with 5min TTL
- Maintainability: separate concerns for model info vs API config
- Adds invalidate_uapi_cache() for config changes

2026-05-30 12:19:42 +08:00

..

__init__.py

bugfix

2025-08-01 21:09:41 +08:00

accounting.py

refactor: add get_llmage_llm() for non-API-call llm data access, replace get_llm() in accounting and pricing

2026-05-29 17:41:43 +08:00

asyncinference.py

bugfix

2026-05-14 15:19:13 +08:00

callback.py

bugfix

2026-04-08 17:23:24 +08:00

init.py

refactor: get_llm() uses get_llmage_llm() + cached uapi/uapiio lookups

2026-05-30 12:19:42 +08:00

jimeng.py

bugfix

2025-09-23 12:15:24 +08:00

keling.py

bugfix

2025-09-25 17:40:00 +08:00

llmclient.py

bugfix

2026-05-15 13:23:10 +08:00

messages.py

bugfix

2025-08-25 11:16:49 +08:00

stats.py

feat: add llmage module stat cards - model count, today's usage, amount, catalog count

2026-05-25 18:48:09 +08:00

syncinference.py

bugfix

2026-05-15 13:23:10 +08:00

utils.py

refactor: get_llm() uses get_llmage_llm() + cached uapi/uapiio lookups

2026-05-30 12:19:42 +08:00