776 commits

Author	SHA1	Message	Date
ishaan-jaff	3a4e512a75	(fix) palm: streaming	2023-12-04 15:06:52 -08:00
Krrish Dholakia	728b879c33	fix(utils.py): fix azure streaming bug	2023-12-04 12:38:22 -08:00
Krrish Dholakia	63e55f1865	fix(proxy_server.py): fix /key/generate post endpoint	2023-12-04 10:44:13 -08:00
ishaan-jaff	93f5c266da	(test) test completion: if 'user' passed to API	2023-12-04 09:50:36 -08:00
Krrish Dholakia	add4dfc528	fix(proxy_server.py): support model info augmenting for azure models	2023-12-02 21:33:54 -08:00
Krrish Dholakia	f72dd24ab9	fix(main.py): set user to none if not passed in	2023-12-02 20:08:25 -08:00
Krrish Dholakia	722c325503	fix(proxy_server.py): update db with master key if set, and fix tracking cost for azure models	2023-12-02 15:58:08 -08:00
Krrish Dholakia	368fee224e	feat: support for azure key vault	2023-12-01 19:36:06 -08:00
Krrish Dholakia	aa36bd2784	fix(utils.py): expand openai_token_counter selection	2023-11-30 18:51:51 -08:00
Krrish Dholakia	ff4457e2d2	fix(router.py): back-off if no models available	2023-11-30 18:42:29 -08:00
Krrish Dholakia	7f04758bcb	(fix) support counting tokens for tool calls	2023-11-30 18:24:21 -08:00
Krrish Dholakia	b4b7acdb72	fix(utils.py): fix azure completion cost calculation	2023-11-30 09:19:35 -08:00
Krrish Dholakia	01c7e18f31	fix(utils.py): include system fingerprint in streaming response object	2023-11-30 08:45:52 -08:00
Krrish Dholakia	e98fac66da	fix(utils.py): fix register model cost map	2023-11-29 21:12:29 -08:00
Krrish Dholakia	0d200cd8dc	feat(main.py): allow updating model cost via completion()	2023-11-29 20:14:39 -08:00
ishaan-jaff	09caab549a	(feat) async embeddings: OpenAI	2023-11-29 19:35:08 -08:00
ishaan-jaff	2c74dbed17	(chore) util: remove_model_id	2023-11-29 17:30:33 -08:00
Krrish Dholakia	5411d5a6fd	fix(utils.py): raise stop iteration exception on bedrock stream close	2023-11-29 16:43:11 -08:00
Krrish Dholakia	ab76daa90b	fix(bedrock.py): support ai21 / bedrock streaming	2023-11-29 16:35:06 -08:00
ishaan-jaff	2bbd9c063d	(fix) OpenAI embedding	2023-11-29 16:09:31 -08:00
Krrish Dholakia	96a27ce954	fix(utils.py): stop sequence filtering for amazon titan models	2023-11-29 16:04:14 -08:00
Krrish Dholakia	6c98715b94	bump: version 1.7.13 → 1.7.14	2023-11-29 15:19:18 -08:00
Krrish Dholakia	2b437a2699	fix(utils.py): return last streaming chunk	2023-11-29 12:11:08 -08:00
Krrish Dholakia	b6bc75e27a	fix(utils.py): fix parallel tool calling when streaming	2023-11-29 10:56:21 -08:00
Krrish Dholakia	9024a47dc2	fix(utils.py): bedrock/cohere optional params	2023-11-29 08:08:48 -08:00
Krrish Dholakia	4e9aa0d338	fix(utils.py): fix bedrock/cohere supported params	2023-11-28 17:42:50 -08:00
Krrish Dholakia	bb1267eb07	fix(router.py): fix exponential backoff to use retry-after if present in headers	2023-11-28 17:25:03 -08:00
Krrish Dholakia	5ed957ebbe	fix(utils.py): bug fix return only non-null responses	2023-11-28 09:43:42 -08:00
Krrish Dholakia	e8331a4647	fix(utils.py): azure tool calling streaming	2023-11-27 19:07:38 -08:00
Krrish Dholakia	4cdd930fa2	fix(stream_chunk_builder): adding support for tool calling in completion counting	2023-11-27 18:39:47 -08:00
ishaan-jaff	9cef551623	(feat) raise APIConnectionError error for Azure +OpenAI	2023-11-27 18:08:47 -08:00
Krrish Dholakia	04f745e314	fix(router.py): speed improvements to the router	2023-11-27 17:35:26 -08:00
ishaan-jaff	d0538e32c9	(feat) completion: sagemaker debugging - show boto3 request sent	2023-11-27 09:04:50 -08:00
Krrish Dholakia	59ba1560e5	fix(router.py): fix fallbacks	2023-11-25 19:34:20 -08:00
Krrish Dholakia	ab0bc87427	fix(router.py): check if fallbacks is none	2023-11-25 14:58:07 -08:00
Krrish Dholakia	95579fda7d	fix(utils.py): fix bedrock + cohere calls	2023-11-25 14:45:42 -08:00
Krrish Dholakia	2eb7386095	fix(utils.py): fix embedding response object conversion	2023-11-25 14:25:06 -08:00
Krrish Dholakia	e43e1e9ab1	fix(utils.py): fix linting errors	2023-11-25 14:19:14 -08:00
Krrish Dholakia	ec81b393e2	fix(utils.py): fix linting erros	2023-11-25 14:14:46 -08:00
Krrish Dholakia	a070bee5be	fix(utils.py): fix linting issues	2023-11-25 14:11:40 -08:00
Krrish Dholakia	8970f12780	fix(utils.py): fix linting issues	2023-11-25 13:48:50 -08:00
Krrish Dholakia	6d9f7b8f9d	fix: fix nlp cloud streaming	2023-11-25 13:45:23 -08:00
Krrish Dholakia	30f47d3169	bump: version 1.7.0 → 1.7.1	2023-11-25 12:34:28 -08:00
Krrish Dholakia	dac76a4861	fix(utils.py): fix embedding response output parsing	2023-11-25 12:06:57 -08:00
Krrish Dholakia	e732fb8b97	fix(main.py): logit bias mapping for batch_completions	2023-11-24 16:05:51 -08:00
ishaan-jaff	0f873c756d	(fix) completion: OpenAI/Azure filter out None params	2023-11-24 14:01:21 -08:00
Krrish Dholakia	4a5dae3941	fix(main.py): fix streaming_chunk_builder to return usage	2023-11-24 11:27:04 -08:00
ishaan-jaff	19fb24cd15	(feat) cost tracking for azure llms	2023-11-23 21:41:38 -08:00
Krrish Dholakia	7d221fe863	fix(utils.py): make failure logging sync	2023-11-23 20:19:27 -08:00
ishaan-jaff	695eaac542	(fix) cost calculator for FT: gpt-3.5	2023-11-23 18:28:21 -08:00