scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-10 22:41:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 82
25011 commits 2289 branches 1405 tags 2.4 GiB
Commit graph

25011 commits

This branch
This branch
All branches
Author SHA1 Message Date
Tim Elfrink
9d7942eb35
Fix: Vertex AI Gemini labels field provider-aware filtering (#14563) 
* Add comprehensive tests for Vertex AI Gemini labels provider filtering

- Test Google GenAI endpoints exclude labels even when explicitly provided
- Test Vertex AI endpoints include labels when provided
- Cover provider detection logic for different endpoint URLs
- Verify metadata-to-labels conversion only happens for Vertex AI
- Ensure edge cases are handled properly (null/empty api_base)

* Fix Vertex AI Gemini labels field provider-aware filtering

- Add _is_google_genai_endpoint() function to detect Google GenAI vs Vertex AI endpoints
- Update _transform_request_body() to accept api_base parameter
- Only include labels field for Vertex AI endpoints (not Google GenAI)
- Pass api_base through sync/async transform functions
- Maintain backward compatibility with existing usage
- Fixes issue where Google GenAI requests failed with unsupported labels field

* Refactor labels filtering to use custom_llm_provider instead of URL parsing

Replace URL-based endpoint detection with custom_llm_provider parameter
checking for cleaner, more reliable provider identification.

Changes:
- Remove _is_google_genai_endpoint() helper function
- Update labels condition to use custom_llm_provider != "gemini"
- Remove api_base parameter from _transform_request_body()
- Simplify sync/async transform function signatures
- Update tests to reflect new parameter structure
- Remove obsolete test_provider_detection test

This approach aligns with existing codebase patterns where
custom_llm_provider="gemini" identifies Google AI Studio endpoints
that don't support labels, while vertex_ai/vertex_ai_beta identify
Vertex AI endpoints that do support labels.

* Use LlmProviders.GEMINI constant instead of hardcoded string
 2025-09-15 12:43:07 -07:00
Mubashir Osmani
321d5299b2
s3_endpoint_url returned 404 (#14559) 
* added spend metrics

* feat: Add Spend metrics in datadog

* fix: lint errors

* fix: s3 endpoint url logging

* fixed lint errors

* remove from branch

This reverts commit e123cae06e.

* Remove from branch

This reverts commit e694cc102a.

* remove "added spend metrics"

This reverts commit 6156590190.
 2025-09-15 12:08:18 -07:00
Ishaan Jaff
cebacd65cf
[Bug Fix] SCIM v2 - ensure group PUSH and PUT ops allow creating non-existent members (#14581) 
* fix: scim handle non existent members

* test - scim v2

* test fix

* fix: NewUserResponse
 2025-09-15 11:27:05 -07:00
Tim Elfrink
30c3e7b3d3
Fix: Bedrock cross-region inference profile cost calculation (#14566) 
* Add tests for Bedrock cross-region inference profile mapping

- Test model mapping lookup works correctly
- Test proxy cost calculation scenario reproduces original issue
- Verify cost calculation returns expected values
- Ensure compatibility with existing test patterns

* Fix Bedrock cross-region inference profile cost calculation

- Add mapping for bedrock/us.anthropic.claude-3-5-haiku-20241022-v1:0
- Sync backup file for local testing consistency
- Resolve proxy spend tracking failures for cross-region profiles
- Maintain identical configuration with standalone profile

Fixes #14458
 2025-09-15 07:10:20 -07:00
Sameer Kankute
110ce543c2
[Feat]Add cancel endpoint support for openai and azure (#14561) 
* Add cancel endpoint support for openai
 and azure

* fix lint error

* fix cancel url contruction azure

* readd changes
 2025-09-15 07:08:56 -07:00
Sameer Kankute
7fd6e62570
Fix unsupported stop param for grok-code models (#14565) 2025-09-15 07:00:54 -07:00
Krrish Dholakia
03f2be1e20 fix: fix race conditions 2025-09-14 09:41:04 -07:00
Krrish Dholakia
2c6481fa33 fix: remove incorrect test 2025-09-14 09:34:31 -07:00
Krrish Dholakia
fc2d1f2646 fix: fix import errors 2025-09-14 09:32:21 -07:00
Krish Dholakia
510332b886
Merge pull request #14491 from Rasmusafj/main 
Resolve cache key collision issue where all soft budget alerts use identical cache keys
 2025-09-14 00:51:27 -07:00
Krish Dholakia
56fd60b140
Merge pull request #14494 from eliasto/feat/ovhcloud-ai-edpoints-provider 
feat: Add OVHCloud AI Endpoints as a provider
 2025-09-14 00:45:08 -07:00
Krish Dholakia
37aa8fff5e
Merge pull request #14548 from nearai-cloud/fix/completion-chat-id 
fix: completion chat id
 2025-09-14 00:42:04 -07:00
Krrish Dholakia
7ef8c808cf docs: update doc 2025-09-14 00:28:11 -07:00
Krish Dholakia
db644b6edd
Merge pull request #14500 from luisfucros/feat/update-sambanova-models 
Add sambanova deepseek v3.1 and gpt-oss-120b models
 2025-09-13 23:38:20 -07:00
Krish Dholakia
11822e63f1
Merge pull request #14519 from uc4w6c/feat/add_tools_permission_guardrail 
feat: add tool-permission guardrail
 2025-09-13 23:22:31 -07:00
Coffee
ac0386ae1f fix: completion chat id 2025-09-14 14:21:57 +08:00
Krish Dholakia
dc4bbba0a5
Merge pull request #14520 from boopesh07/email_prometheus 
Added user_email labels to the prometheus monitoring.
 2025-09-13 23:19:46 -07:00
Krish Dholakia
2ec4b2953c
Merge pull request #14531 from mubashir1osmani/main 
fix: DD tool calls passed in metadata
 2025-09-13 23:16:15 -07:00
Krish Dholakia
2338dd952e
Merge pull request #14546 from BerriAI/filter-on-logs-bug 
The 'last 24 hours' button shows up above the end user dropdown on Logs page
 2025-09-13 23:12:39 -07:00
Krish Dholakia
1622d03ecc
Merge pull request #14545 from BerriAI/litellm_ui_qa_09_13_2025_p1 
Litellm UI qa 09 13 2025 p1 - fix end user filtering + fix load mcp tool call error + prevent setting max user budget on scroll in edit user settings
 2025-09-13 18:48:43 -07:00
Krrish Dholakia
fe546b936a fix(fetch_mcp_tools.tsx): fix load mcp tools 2025-09-13 18:42:42 -07:00
Ishaan Jaff
f37dd6bb95
Litellm 1.77.2 stable notes (#14544) 
* fix release notes instructions

* docs v1

* fix doc

* fix highlights

* docs fix

* docs fix
 2025-09-13 18:41:34 -07:00
Krrish Dholakia
84f934bf36 fix(user_edit_view.tsx): use shared component to prevent accidental setting of edit user budget 2025-09-13 18:20:07 -07:00
Krrish Dholakia
bb5e71447d fix(spend_management_endpoints.py): add end user filtering 2025-09-13 18:15:34 -07:00
Boopesh Shanmugam
95da14cb96
Docs update on user header mapping (#14527) 2025-09-13 16:58:10 -07:00
Ishaan Jaff
6172145014
fix: org budget update fix (#14541) 2025-09-13 16:34:26 -07:00
Krrish Dholakia
b6bca6369f fix(constants.py): make default num workers 1 2025-09-13 13:57:58 -07:00
Ishaan Jaff
9110af37d0 bump: version 1.77.1 → 1.77.2 2025-09-13 13:55:30 -07:00
Krrish Dholakia
dc4b09e26e build(ui/): new ui build 2025-09-13 13:48:27 -07:00
Krish Dholakia
6efc898407
Merge pull request #14523 from BerriAI/litellm_dev_09_12_2025_p1 
VLLM - transcription endpoint support + Ollama_chat/ - images, thinking, and content as list handling +
 2025-09-13 13:39:28 -07:00
Ishaan Jaff
6c27e5ce43 fix: DEFAULT_NUM_WORKERS_LITELLM_PROXY 2025-09-13 12:11:33 -07:00
Ishaan Jaff
252ec8e1ae test_normal_router_call_tpm_v3 2025-09-13 12:04:56 -07:00
Alexsander Hamir
44d209622b
fix: remove dynamic creation of static value (#14538) 2025-09-13 11:58:38 -07:00
Krrish Dholakia
8443000ca4 fix(main.py): route vllm calls via the openai sdk route 
consistent with other openai-like implementations
 2025-09-13 11:49:05 -07:00
Ishaan Jaff
26dafdc493 test fix: note this does not play nice with circleCI, it passes on local 2025-09-13 11:37:19 -07:00
Ishaan Jaff
ea280e1a6e _transform_messages: google ai studio 2025-09-13 11:31:19 -07:00
dependabot[bot]
d89c7f0eb7
build(deps): bump hono from 4.6.5 to 4.9.7 in /litellm-js/spend-logs (#14513) 
Bumps [hono](https://github.com/honojs/hono) from 4.6.5 to 4.9.7.
- [Release notes](https://github.com/honojs/hono/releases)
- [Commits](https://github.com/honojs/hono/compare/v4.6.5...v4.9.7)

---
updated-dependencies:
- dependency-name: hono
  dependency-version: 4.9.7
  dependency-type: direct:production
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
 2025-09-13 11:10:37 -07:00
Krrish Dholakia
a8e2d24d3a fix: fix import 2025-09-13 10:12:32 -07:00
Krish Dholakia
269515e525
Merge branch 'main' into litellm_dev_09_12_2025_p1 2025-09-13 10:10:30 -07:00
Ishaan Jaff
ec32a0a7d7 fix: get_base_completion_call_args 2025-09-13 09:59:08 -07:00
Krrish Dholakia
9ee9745781 docs(vllm.md): document new endpoint 2025-09-13 09:46:39 -07:00
Ishaan Jaff
69e451458c test_logging_opentelemetry_context_propagation 2025-09-13 09:44:36 -07:00
Krish Dholakia
38efd358eb
Merge pull request #14401 from Noma-Security/noma_non_blocking_monitor_mode 
Noma non blocking monitor mode & anonymize input support
 2025-09-13 09:41:41 -07:00
Krish Dholakia
550feffeb9
Merge pull request #14512 from timelfrink/fix/lm-studio-bearer-header-14502 
fix(lm_studio): resolve illegal Bearer header value issue
 2025-09-13 09:40:30 -07:00
Krish Dholakia
a6ba744085
Merge pull request #14529 from BerriAI/litellm_contributor_prs_09_12_2025_p1 
Fixed Log Tab Key Alias
 2025-09-13 09:39:28 -07:00
Krish Dholakia
ad9515a81e
Merge branch 'main' into litellm_contributor_prs_09_12_2025_p1 2025-09-13 09:38:43 -07:00
mubashir1osmani
3cca2e95a7 fix lint errors 2025-09-13 00:17:57 -04:00
mubashir1osmani
fd11159148 fix: DD tool calls passed in metadata 2025-09-13 00:11:18 -04:00
Krish Dholakia
de8cf40ffa
Merge pull request #14482 from BerriAI/litellm_dev_09_11_2025_p3 
Feature - new `litellm_request_debug=true` flag, enables emitting raw request/response log on single request
 2025-09-12 19:56:53 -07:00
Krish Dholakia
663dbc6080
Merge pull request #14477 from BerriAI/litellm_dev_09_11_2025_p2 
`/v1/messages` - don't send content block after message w/ finish reason + usage block + `/key/unblock` - support hashed tokens
 2025-09-12 19:51:44 -07:00