Krrish Dholakia
691c185ff8
feat(router.py): support region routing for bedrock, vertex ai, watsonx
2024-05-11 11:04:00 -07:00
Krrish Dholakia
2ed155b4d4
feat(router.py): allow setting model_region in litellm_params
...
Closes https://github.com/BerriAI/litellm/issues/3580
2024-05-11 10:18:08 -07:00
Ishaan Jaff
b02f633cd6
Merge pull request #3577 from BerriAI/litellm_add_triton_server
...
[Feat] Add Triton Embeddings to LiteLLM
2024-05-10 19:20:23 -07:00
Ishaan Jaff
82344db621
fix triton params
2024-05-10 19:14:48 -07:00
Krish Dholakia
859d978a77
Merge pull request #3571 from BerriAI/litellm_hf_classifier_support
...
Huggingface classifier support
2024-05-10 17:54:27 -07:00
Ishaan Jaff
64c9871583
fix langfuse logger re-initialized on all failure callbacks
2024-05-10 17:48:44 -07:00
Ishaan Jaff
4c0a1d3ec6
fix langfuse failure logging
2024-05-10 17:02:38 -07:00
Ishaan Jaff
472ad0b800
fix - support dynamic failure callbacks
2024-05-10 16:37:01 -07:00
Ishaan Jaff
b0777de041
fix - using failure callbacks with team based logging
2024-05-10 16:18:13 -07:00
Krrish Dholakia
4680f4e1db
test(test_completion.py): reintegrate testing for huggingface tgi + non-tgi
2024-05-10 14:07:01 -07:00
Krrish Dholakia
03139e1769
fix(main.py): support env var 'VERTEX_PROJECT' and 'VERTEX_LOCATION'
2024-05-10 07:57:56 -07:00
Krish Dholakia
ddf09a3193
Merge pull request #3552 from BerriAI/litellm_predibase_support
...
feat(predibase.py): add support for predibase provider
2024-05-09 22:21:16 -07:00
Ishaan Jaff
a9aa71de01
Merge pull request #3547 from BerriAI/litellm_support_stream_options_text_completion
...
[Feat] support `stream_options` on `litellm.text_completion`
2024-05-09 18:05:58 -07:00
Krrish Dholakia
7c0ab40fd5
feat(predibase.py): support async_completion + streaming (sync + async)
...
finishes up pr
2024-05-09 17:41:27 -07:00
Krrish Dholakia
f660d21743
feat(predibase.py): add support for predibase provider
...
Closes https://github.com/BerriAI/litellm/issues/1253
2024-05-09 16:39:43 -07:00
Krrish Dholakia
f10413e373
fix(utils.py): change error log to be debug
2024-05-09 13:58:45 -07:00
Ishaan Jaff
b0bcb74ba5
fix TextCompletionStreamWrapper
2024-05-09 09:54:44 -07:00
Ishaan Jaff
454dbdf285
feat - support stream_options for text completion
2024-05-09 08:42:25 -07:00
Ishaan Jaff
2968737969
Merge pull request #3537 from BerriAI/litellm_support_stream_options_param
...
[Feat] support `stream_options` param for OpenAI
2024-05-09 08:34:08 -07:00
Krrish Dholakia
40b1ee42ed
fix(get_api_base): fix get_api_base to handle model with alias
2024-05-09 08:01:17 -07:00
Krish Dholakia
8af4596dad
Revert "Add support for async streaming to watsonx provider "
2024-05-09 07:44:15 -07:00
Krish Dholakia
64ca2fde53
Merge branch 'main' into litellm_region_based_routing
2024-05-08 22:19:51 -07:00
Krish Dholakia
ffe255ea2b
Merge pull request #3479 from simonsanvil/feature/watsonx-integration
...
Add support for async streaming to watsonx provider
2024-05-08 22:19:05 -07:00
Krrish Dholakia
0ea8222508
feat(router.py): enable filtering model group by 'allowed_model_region'
2024-05-08 22:10:17 -07:00
Ishaan Jaff
8fb55507ad
support stream_options
2024-05-08 21:53:33 -07:00
Ishaan Jaff
fed005d853
Merge pull request #3534 from BerriAI/litellm_fix_cost_calc_bedrock
...
[Fix] `litellm.completion_cost(model="bedrock/anthropic.claude-instant-v1"..)`
2024-05-08 16:59:46 -07:00
Krrish Dholakia
5f93cae3ff
feat(proxy_server.py): return litellm version in response headers
2024-05-08 16:00:08 -07:00
Ishaan Jaff
6d71c1e44b
fix completion cost test
2024-05-08 15:51:30 -07:00
Ishaan Jaff
bbd8770260
fix - cost tracking - looking up bedrock pricing
2024-05-08 15:25:52 -07:00
Ishaan Jaff
62fa6661df
Merge pull request #3439 from phact/patch-3
...
add_function_to_prompt bug fix
2024-05-07 19:31:19 -07:00
phact
9d64d4e84f
locals().copy()
2024-05-07 14:58:35 -04:00
Paul Gauthier
c72e7e85e2
Added support for the deepseek api
2024-05-07 11:44:03 -07:00
phact
ada5f69ccd
looks like cohere does support function calling
2024-05-07 13:41:05 -04:00
phact
bbebfa5af0
unit test and list fix
2024-05-07 13:24:28 -04:00
Krish Dholakia
72736599db
Merge pull request #3459 from alexanderepstein/langfuse_improvements
...
Update support for langfuse metadata
2024-05-06 21:56:29 -07:00
Krish Dholakia
185581d5a9
Merge branch 'main' into litellm_slack_daily_reports
2024-05-06 19:31:20 -07:00
Krrish Dholakia
0c1fc34e9f
feat(slack_alerting.py): support sending daily reports on deployments
...
allow admin to easily know slow + failing deployments
Closes https://github.com/BerriAI/litellm/issues/3483
2024-05-06 17:18:42 -07:00
Ishaan Jaff
28a05fce3b
fix test router debug logs
2024-05-06 16:38:16 -07:00
Ishaan Jaff
1d814a7234
fix add key name + team name in alerting messages
2024-05-06 14:29:04 -07:00
Krrish Dholakia
f1c2423ca4
fix(utils.py): handle gemini chunk no parts error
...
Fixes https://github.com/BerriAI/litellm/issues/3468
2024-05-06 10:59:53 -07:00
Simon Sanchez Viloria
9a95fa9348
Merge branch 'main' into feature/watsonx-integration
2024-05-06 17:27:14 +02:00
Simon Sanchez Viloria
361188b436
(feat) support for async stream to watsonx provider
2024-05-06 17:08:40 +02:00
Krrish Dholakia
80a1344c55
fix(utils.py): anthropic error handling
2024-05-06 07:25:12 -07:00
Alex Epstein
9ae08a8344
Update support for langfuse metadata
...
- Added ability to set trace release, version, metadata
- Added ability to update fields during a trace continuation
- Added ability to update input and output during a trace continuation
- Wrote new test for verifying metadata is set correctly
- Small improvement to setting secret boolean, prevent unnecessary literal_eval
- Small improvements to langfuse tests
2024-05-05 14:56:36 -04:00
Ishaan Jaff
edf8c9a04f
fix don't return num retries in utils.py
2024-05-04 19:07:28 -07:00
Ishaan Jaff
7100ec0a67
feat return num retries in exceptions
2024-05-04 18:50:38 -07:00
Ishaan Jaff
adf09bdd45
fix add get_first_chars_messages in utils
2024-05-04 12:43:09 -07:00
Ishaan Jaff
bea785646f
litellm map Azure GPT ContentPolicyViolationError
2024-05-04 11:14:47 -07:00
Sebastián Estévez
78890e2f33
add_function_to_prompt bug fix
...
This blows up when there's no "functions" in the dictionary even when tools is present because the inner function executes regardless (does not short circuit).
2024-05-03 23:38:54 -04:00
Krish Dholakia
0d3aa8d323
Merge pull request #3433 from BerriAI/litellm_timeout_fix
...
fix(bedrock.py): convert httpx.timeout to boto3 valid timeout
2024-05-03 18:53:59 -07:00
