Commit graph

1456 commits

Author SHA1 Message Date
ishaan-jaff
cf82cd17e4 (temp) prisma client init logic 2024-01-09 13:00:23 +05:30
Krrish Dholakia
27e52794df fix(proxy_server.py): don't reconnect prisma if already connected 2024-01-09 12:59:57 +05:30
Krrish Dholakia
9673e6042e test(test_router.py): fix router test 2024-01-09 12:59:42 +05:30
ishaan-jaff
d3f9a7df65 (fix) test - deprecated textdavinci003 2024-01-09 12:59:42 +05:30
ishaan-jaff
4791514351 (test) hosted - ollama catch timeouts 2024-01-09 12:59:42 +05:30
ishaan-jaff
77027746ba (feat) litellm.completion - support ollama timeout 2024-01-09 12:59:42 +05:30
Krrish Dholakia
10f76ec36c test: testing fixes 2024-01-09 12:59:42 +05:30
ishaan-jaff
520cd7fa89 (ci/cd) use 3 retries for image generation 2024-01-09 12:59:42 +05:30
ishaan-jaff
bae1323cb5 (ci/cd) pytest skip slow replicate test 2024-01-09 12:59:42 +05:30
Krrish Dholakia
22a900463e fix(ollama.py): use tiktoken as backup for prompt token counting 2024-01-09 12:58:58 +05:30
Krrish Dholakia
9ddcdc4716 feat(lowest_latency.py): support expanded time window for latency based routing
uses a 1hr avg. of latency for deployments, to determine which to route to

https://github.com/BerriAI/litellm/issues/1361
2024-01-09 12:58:58 +05:30
ishaan-jaff
7e9359ecb8 (ci/cd) run again 2024-01-09 12:58:58 +05:30
Krrish Dholakia
8913313bc1 test(test_proxy_startup.py): fix gunicorn test 2024-01-09 12:58:58 +05:30
Krrish Dholakia
834d3362d9 test(test_proxy_startup.py): separate tests 2024-01-09 12:58:58 +05:30
Krrish Dholakia
a37b40cdb9 fix(proxy_server.py): add support for passing in config file via worker_config directly + testing 2024-01-09 12:58:58 +05:30
ishaan-jaff
9b04b1c9ad (test) test key/generate against deployed proxy 2024-01-08 17:25:56 +05:30
ishaan-jaff
02c166138f (test) test deployed endpoint 2024-01-08 17:17:31 +05:30
Ishaan Jaff
a70626d6e9
Merge pull request #1356 from BerriAI/litellm_improve_proxy_logs
[Feat] Improve Proxy Logging
2024-01-08 14:41:01 +05:30
ishaan-jaff
96e8c2b4cf (test) tg - ai - set max_tokens=5, fast test 2024-01-08 14:03:31 +05:30
ishaan-jaff
bf30e8fdb2 (test) router- verbose logs with fallbacks 2024-01-08 14:00:12 +05:30
Krrish Dholakia
6333fbfe56 fix(main.py): support cost calculation for text completion streaming object 2024-01-08 12:41:43 +05:30
Krrish Dholakia
9b46412279 fix(utils.py): fix logging for text completion streaming 2024-01-08 12:05:28 +05:30
Krrish Dholakia
b1fd0a164b fix(huggingface_restapi.py): support timeouts for huggingface + openai text completions
https://github.com/BerriAI/litellm/issues/1334
2024-01-08 11:40:56 +05:30
Krrish Dholakia
c720870f80 docs(gemini.md,-deploy.md): doc updates 2024-01-08 11:02:12 +05:30
Krish Dholakia
4ea3e778f7
Merge pull request #1315 from spdustin/feature_allow_claude_prefill
Adds "pre-fill" support for Claude
2024-01-08 10:48:15 +05:30
Krrish Dholakia
3469b5b911 fix(utils.py): map optional params for gemini 2024-01-08 07:38:55 +05:30
Krrish Dholakia
79264b0dab fix(gemini.py): better error handling 2024-01-08 07:32:26 +05:30
Krrish Dholakia
888e21e8e7 test(test_google_ai_studio_gemini.py): use an image url that will work on ci/cd 2024-01-06 22:58:37 +05:30
Krrish Dholakia
1507217725 fix(factory.py): more logging around the image loading for gemini 2024-01-06 22:50:44 +05:30
Krrish Dholakia
5fd2f945f3 fix(factory.py): support gemini-pro-vision on google ai studio
https://github.com/BerriAI/litellm/issues/1329
2024-01-06 22:36:22 +05:30
ishaan-jaff
edac4130bb (fix) s3 + os.environ/ cache test 2024-01-06 16:33:29 +05:30
ishaan-jaff
174248fc71 (test) add back test for counting stream completion tokens 2024-01-06 16:08:32 +05:30
Krish Dholakia
8d32f08858
Merge pull request #1342 from BerriAI/litellm_dockerfile_updates
build(Dockerfile): moves prisma logic to dockerfile
2024-01-06 16:03:25 +05:30
ishaan-jaff
f999b63d05 (test) using os.environ/ on cache + proxy 2024-01-06 15:54:50 +05:30
ishaan-jaff
c2b061acb2 (feat) cache+proxy - set os.environ/ on proxy config 2024-01-06 15:54:16 +05:30
ishaan-jaff
0d152b3748 (fix) cloudflare tests 2024-01-06 15:35:49 +05:30
Krrish Dholakia
9375570547 test(test_async_fn.py): skip cloudflare test - flaky 2024-01-06 15:17:42 +05:30
Krrish Dholakia
04c04d62e3 test(test_stream_chunk_builder.py): remove completion assert, the test is for prompt tokens 2024-01-06 14:12:44 +05:30
Krrish Dholakia
5c45e69a5e test(test_proxy_server_keys.py): add logic for connecting/disconnecting from http server 2024-01-06 14:09:10 +05:30
ishaan-jaff
4a076350cc (ci/cd) move to old version of test_proxy_server_keys.py 2024-01-06 13:03:12 +05:30
ishaan-jaff
41bfd43a48 (ci/cd) pin anyio / async dependencies 2024-01-06 12:38:56 +05:30
ishaan-jaff
3bb49447bc (ci/cd) fix event loop bug proxy_test 2024-01-06 12:30:43 +05:30
ishaan-jaff
79fd2380bb (ci/cd) run again 2024-01-06 12:11:31 +05:30
ishaan-jaff
0ebd0653c5 (ci/cd) make prisma tests async 2024-01-06 11:43:23 +05:30
spdustin@gmail.com
6201ab2c21 Update factory (and tests) for Claude 2.1 via Bedrock 2024-01-05 23:32:32 +00:00
spdustin@gmail.com
5d074f5b56 Adds tests and updates docs for Claude "pre-fill" 2024-01-05 22:58:41 +00:00
ishaan-jaff
ae54e6d8b0 (ci/cd) proxy:test_add_new_key 2024-01-05 22:53:03 +05:30
ishaan-jaff
40aaac69cc (ci/cd) add print_verbose for /key/generate 2024-01-05 22:38:46 +05:30
ishaan-jaff
dfdd329ddf (ci/cd) pytest event loop fixture 2024-01-05 22:28:34 +05:30
ishaan-jaff
050c289ed1 (ci/cd) test fixture 2024-01-05 22:15:08 +05:30