Commit Graph

59 Commits

Author SHA1 Message Date
Shuchang Zheng
c158ad3f21 migrate observer to task v2 (#1564) 2025-01-15 09:59:18 -08:00
Jason
e54977ef29 feat: Add new model provider Novita AI (#1508) 2025-01-11 00:08:37 -05:00
Shuchang Zheng
5796de73d1 Add AI suggestion endpoints (#1519) 2025-01-08 21:45:38 -08:00
LawyZheng
6b4b52a6c4 hash long href link (#1500) 2025-01-06 23:26:07 +08:00
Shuchang Zheng
175ce55f06 recreate index for artifacts table (#1455) 2024-12-31 01:38:07 -08:00
Shuchang Zheng
b2fdb187fa Use json_repair (#1453) 2024-12-30 14:31:29 -05:00
Shuchang Zheng
5842bfc1fd create cruise related artifact in cruise api (#1355) 2024-12-08 21:17:58 -08:00
Nick Fisher
f7f5c47c8e rename GEMINI_FLUSH->GEMINI_FLASH (#1333) 2024-12-05 20:28:54 -08:00
Shuchang Zheng
d8c6471c1b amazon nova support (#1315) 2024-12-05 00:12:53 -08:00
Shuchang Zheng
6db01f4783 gemini 1.5 pro & flash (#1308) 2024-12-02 15:59:59 -08:00
Shuchang Zheng
57e6404881 add gemeni pro support (#1307) 2024-12-02 15:51:18 -08:00
Shuchang Zheng
7f6b2c0929 shu/removeSettingsManager.get_settings (#1305) 2024-12-02 15:01:22 -08:00
Shuchang Zheng
675933128c support bedrock sonnet 3.5 inference profile (#1195) 2024-11-14 14:37:53 -08:00
Shuchang Zheng
281feb60e4 Remove Call LLM API Log (#1132) 2024-11-04 22:00:15 -05:00
Shuchang Zheng
2ac8a1a7d0 SKY-3915 Remove incomplete json fixing logic -- models have gotten better (#1121) 2024-11-04 00:56:58 -05:00
Shuchang Zheng
bf06fcfeb7 Update max output tokens to 16K (#1110) 2024-11-02 21:46:55 -07:00
Suchintan
7cba401e2e Add support for the latest sonnet 3.5 (#1025) 2024-10-22 16:57:26 -04:00
Shuchang Zheng
60069a6256 fix LiteLLMParams (#958) 2024-10-10 23:24:27 -07:00
Shuchang Zheng
b536f1d27f gpt4o mini pricing for massive (#957) 2024-10-10 22:57:54 -07:00
Shuchang Zheng
06daa9834f catch litellm no deployment error (token ratelimit / ValueError) and raise LLMProviderErrorRetryableTask (#758) 2024-08-29 12:19:51 -07:00
Shuchang Zheng
b5d743d6eb handle litellm APIError for the router and raise LLMProviderErrorRetryableTask (#750) 2024-08-28 10:23:16 -07:00
Kerem Yilmaz
a6a545224a adding step_exception to provide more signal of why the step failed (#730) 2024-08-28 01:06:04 -07:00
Shuchang Zheng
1967e44acc add support for gpt-4o-2024-08-06 (#715) 2024-08-22 14:08:43 -07:00
LawyZheng
46a139bd27 litellm cooldown config (#711) 2024-08-19 16:49:26 +08:00
LawyZheng
98e2f7f206 add gpt4o mini support (#666) 2024-08-02 19:35:52 +08:00
Shuchang Zheng
dd188b5f4d fix t_llm_request is unbounded problem (#657) 2024-07-29 13:54:42 -07:00
Shuchang Zheng
c52f8f2433 shu/making scrolling debug logs (#635) 2024-07-25 12:56:10 -07:00
Kerem Yilmaz
20a611a935 Use perf-counter instead of time (#643) 2024-07-24 23:47:02 -07:00
Kerem Yilmaz
5000f3d762 Catch CanceledError in litellm call (#642) 2024-07-24 23:41:34 -07:00
Shuchang Zheng
ec5a0a03c0 add gpt4o-mini support (#625) 2024-07-18 22:33:19 -07:00
Kerem Yilmaz
fd9419b33a clean up gpt4 routers. add model_info base_model for azure gpt router (#620) 2024-07-18 15:37:20 -07:00
Kerem Yilmaz
0ab41a207a Start using Claude 3.5 Sonnet (#497) 2024-06-20 10:33:17 -07:00
Kerem Yilmaz
14009bae34 Ykeremy/gpt4o router (#431) 2024-06-05 21:36:32 -07:00
Kerem Yilmaz
04c6e55848 Keep track of token counts in steps table (#412) 2024-06-03 15:55:34 -07:00
Kerem Yilmaz
753bac3174 Set 5 min timeout for LLM calls (#358) 2024-05-23 18:49:49 -07:00
Kerem Yilmaz
df2aee3c79 Add logs (#357) 2024-05-23 12:07:49 -07:00
Kerem Yilmaz
e6d4302d8c Auto-fix invalid JSON (#354) 2024-05-21 22:04:32 -07:00
Kerem Yilmaz
6908ca5844 Pass base parameters to acompletion (#343) 2024-05-17 16:36:15 -07:00
Suchintan
9a6ec13af6 Ruff (#339) 2024-05-17 12:10:54 -04:00
Divyansh Verma
8598f56a83 GPT4 integration (#324)
Co-authored-by: Suchintan <suchintan@users.noreply.github.com>
2024-05-17 11:25:03 -04:00
Hasnain
45a2840538 gpt-4o integration (#333) 2024-05-17 11:22:08 -04:00
Shuchang Zheng
2466897158 adopt ruff as the replacement for python black (#332) 2024-05-16 18:20:11 -07:00
Shuchang Zheng
9e1b07451f update pre-commit (#330) 2024-05-16 17:11:49 -07:00
OB42
72389b217c add missing check for ENABLE_BEDROCK in LLMConfigRegistry (#298)
Co-authored-by: Kerem Yilmaz <kerem@skyvern.com>
2024-05-14 10:07:47 -07:00
Kerem Yilmaz
cf01e81ba2 Force Claude 3 models to output JSON object and parse it more reliably (#293)
Co-authored-by: otmane <otmanebenazzou.pro@gmail.com>
2024-05-10 00:51:12 -07:00
Kerem Yilmaz
bf55bf8314 Default the old anthropic model name for backwards compatibility (#285) 2024-05-08 23:43:52 -04:00
Kerem Yilmaz
e5d094493e Define haiku + prompt engine takes a directory arg (#279) 2024-05-08 02:07:18 -07:00
Kerem Yilmaz
4459ca7d09 Integrate posthog feature flags + setup Claude 3 experiment (#251) 2024-05-02 15:06:21 -07:00
Shuchang Zheng
a1fa5ee4e0 log main_model_group when router.acompletion has an unexpected error (#249) 2024-05-02 14:20:49 -04:00
Kerem Yilmaz
3d712cb0fb llm router for gpt4 turbo (#237) 2024-04-25 01:32:22 -07:00