Commit Graph
36424 Commits
Author SHA1 Message Date
Krrish Dholakia a671275f5c ci: add zizmor github action 2026-03-27 05:33:21 -07:00
Sameer Kankute 226e18534e Add correct pricing 2026-03-27 11:04:29 +05:30
Sameer Kankute 06c8476544 feat(gemini): add gemini-3.1-flash-live-preview to model cost map
Made-with: Cursor
2026-03-27 11:04:29 +05:30
Krrish Dholakia dfb543369b fix: address zizmor comments 2026-03-26 21:09:01 -07:00
yuneng-jiangandGitHub b20cff8673 Merge pull request #24654 from BerriAI/litellm_publish_to_pypi_migrate
[Infra] Migrate PyPI Publishing from CircleCI to GitHub Actions OIDC
2026-03-26 19:09:00 -07:00
Yuneng Jiang ba8455a3be [Infra] Migrate PyPI publishing from CircleCI to GitHub Actions OIDC
- Add .github/workflows/publish_to_pypi.yml with OIDC trusted publisher
- Remove publish_to_pypi job from .circleci/config.yml
- Zero long-lived tokens, all actions SHA-pinned, build deps version-pinned
2026-03-26 19:02:14 -07:00
Nicholas Gigliotti 92654bad37 Refactor _supports_native_structured_outputs to use standard supports_* utility pattern
Addresses Greptile review feedback: replace direct litellm.model_cost
lookup with the standard _supports_factory infrastructure used by
supports_reasoning, supports_native_streaming, etc.

- Add supports_native_structured_output() utility in litellm/utils.py
- Add supports_native_structured_output field to ModelInfoBase type
- Wire field into _get_model_info_helper return dict
- Delegate from Bedrock _supports_native_structured_outputs to utility
- Add field to JSON schema validator in test_utils.py
2026-03-26 21:49:03 -04:00
Nicholas Gigliotti b45ef7f6b9 Keep multimodal gemini-embedding-2-preview entry to align with #23599
Main has two duplicate keys for vertex_ai/gemini-embedding-2-preview.
Our JSON round-trip collapsed them to the second (text-only) entry, but
PR #23599 intentionally keeps the first (multimodal pricing) entry.
Restore the multimodal entry to avoid conflicts.
2026-03-26 20:23:12 -04:00
Nicholas Gigliotti 0ef8eb6121 Add test assertion for deepseek.v3-v1:0 native structured output 2026-03-26 20:23:12 -04:00
Nicholas Gigliotti d7e55bf105 Fix test state leakage: restore env and model_cost after each test
Wrap cost-map-dependent tests in try/finally to restore
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] and litellm.model_cost,
preventing test-ordering sensitivity.
2026-03-26 20:23:12 -04:00
Nicholas Gigliotti ed06dded19 Remove native structured output flag from minimax-m2.1, kimi-k2.5, qwen3-coder-next
Integration testing confirmed:
- minimax.minimax-m2.1: Bedrock rejects outputConfig.textFormat (400)
- moonshotai.kimi-k2.5: Bedrock rejects outputConfig.textFormat (400)
- qwen.qwen3-coder-next: unavailable in us-east-1 and us-west-2
2026-03-26 20:23:12 -04:00
Nicholas Gigliotti a7ebc72c26 Add native structured output flag for deepseek.v3-v1:0
Integration tested 28/28 (10 sync + 10 streaming + extras) on the
native outputConfig.textFormat path in us-west-2. deepseek.v3.2 does
not support native structured output (Bedrock returns 400).
2026-03-26 20:23:12 -04:00
Nicholas Gigliotti aba027beed Remove native structured output flag from models broken on Bedrock
Integration testing confirmed gemma-3 (4b/12b/27b) ignores the JSON
schema and returns free text, and nemotron-nano (9b/12b) errors with
"Tool calling is not supported in streaming mode" even on sync calls.
Remove the flag so these models fall back to the tool-call approach.
Also fix test assertions to match (nemotron-nano-3-30b is supported,
gemma-3 and nemotron-nano-12b are not).
2026-03-26 20:23:12 -04:00
Nicholas Gigliotti cb66672017 Replace hardcoded Bedrock native structured output model set with cost JSON lookup
Move the source of truth for which Bedrock models support native structured
outputs (outputConfig.textFormat) from a hardcoded substring set
(BEDROCK_NATIVE_STRUCTURED_OUTPUT_MODELS) to the cost JSON via a new
"supports_native_structured_output" flag. This makes it possible to add
support for new models (including Claude Sonnet 4.6, which was missing)
by updating the JSON alone, with no code changes needed.
2026-03-26 20:23:12 -04:00
Krrish Dholakia efd73a44f4 fix(index.md): test commit 2026-03-26 16:12:00 -07:00
Krrish Dholakia 8f425ec3ff fix: update ioc to include checkmarx.zone 2026-03-26 15:43:01 -07:00
ishaan-berriandGitHub 6153793f81 Merge pull request #24649 from BerriAI/ishaan-berri-patch-1
Update README.md
2026-03-26 15:22:23 -07:00
ishaan-berriandGitHub b69ce9fafa Update README.md 2026-03-26 15:22:11 -07:00
yuneng-jiangandGitHub 25feae9f0f Merge pull request #24607 from BerriAI/litellm_gha_pin_pt_2
[Infra] Pin GHA dependencies and remove unused load test files
2026-03-26 08:47:38 -07:00
yuneng-jiangandGitHub bdf4acc472 Merge pull request #24468 from BerriAI/litellm_opencode-drop-reasoningSummary-docs
docs(opencode): add guidance for dropping reasoningSummary param
2026-03-25 23:00:56 -07:00
Yuneng JiangandClaude Opus 4.6 84be6f69ef fix google-cloud-aiplatform pin to be compatible with google-genai==1.22.0
Pin to 1.115.0 (latest version that doesn't require google-genai>=1.59.0).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-25 22:37:43 -07:00
Yuneng JiangandClaude Opus 4.6 1beb687f54 pin GHA dependencies + remove unused load test files
Pin all pip install commands to exact versions and SHA-pin all GitHub
Actions to prevent supply chain attacks. Remove snok/install-poetry
in favor of direct pip install. Delete orphaned load test scripts.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-03-25 22:32:08 -07:00
Krrish Dholakia df2a36dd27 docs: document new github + gitlab ci scripts 2026-03-25 20:17:10 -07:00
yuneng-jiangandGitHub 437341c9b5 Merge pull request #24594 from BerriAI/litellm_gha_pin_helm
[Infra] Pins GitHub Action workflows to specific hashes
2026-03-25 18:17:21 -07:00
Yuneng Jiang b90a0af0d7 remove extra @ 2026-03-25 17:46:37 -07:00
Yuneng Jiang a989587525 re-add helm unit test with checksum pin 2026-03-25 17:38:36 -07:00
Yuneng Jiang f86b240d7e pin github scripts + remove unused 2026-03-25 17:38:36 -07:00
ryan-crabbe-berriandGitHub 0d2b454cec Merge pull request #24490 from BerriAI/litellm_ryan-march-23
Litellm ryan march 23
2026-03-25 16:04:12 -07:00
ishaan-berriandGitHub 90b850ef8e Merge pull request #24587 from BerriAI/remove-phone-numbers
docs: remove phone numbers from readme and docs
2026-03-25 12:45:42 -07:00
Ishaan Jaffer a2f02aa139 docs: remove phone numbers from readme and docs 2026-03-25 12:40:40 -07:00
ishaan-berriandGitHub 67609e04bc Merge pull request #24584 from BerriAI/worktree-floofy-prancing-crystal
security: remove .claude/settings.json and block re-adding via semgrep
2026-03-25 12:28:06 -07:00
Ishaan Jaffer b77e1ccea6 security: remove .claude/settings.json and add semgrep rule to prevent re-adding 2026-03-25 11:57:43 -07:00
Krrish Dholakia 7d7045cbc1 fix: cleanup 2026-03-25 09:20:25 -07:00
Krrish Dholakia 10a48f7655 docs: add tldr 2026-03-24 19:43:13 -07:00
ishaan-berriandGitHub d93ee444fd Merge pull request #24541 from BerriAI/workflow-deletions
ci: remove publish-migrations and reset_stable workflows
2026-03-24 18:53:33 -07:00
Ishaan Jaffer 3e8a6f24b7 ci: remove all publish/deploy workflows as part of supply chain incident response 2026-03-24 18:03:04 -07:00
Ishaan Jaffer 3d5b1ecb3b ci: remove publish-migrations and reset_stable workflows 2026-03-24 17:59:07 -07:00
ishaan-berriandGitHub b26bc5a91c Merge pull request #24540 from BerriAI/worktree-tingly-jumping-lovelace
docs: minor updates to security update blog post
2026-03-24 17:35:01 -07:00
ishaan-berriandGitHub e92edee7e7 Merge branch 'main' into worktree-tingly-jumping-lovelace 2026-03-24 17:34:51 -07:00
Ishaan Jaffer f9d1f8cde6 docs: minor updates to security update blog post 2026-03-24 17:15:44 -07:00
Krrish Dholakia 4571f2a6c1 fix: add tldr to blog post 2026-03-24 16:54:59 -07:00
ishaan-berriandGitHub 7455b115bf Merge pull request #24537 from BerriAI/litellm_blog-security-incident
docs: add security update blog post for March 2026 supply chain incident
2026-03-24 16:07:45 -07:00
Ryan Crabbe eaa025ef7d docs: add security update blog post for March 2026 supply chain incident 2026-03-24 15:54:35 -07:00
Krrish Dholakia f9d29e4e4e fix(security_scans.sh): move to older trivvy version, based on https://github.com/BerriAI/litellm/issues/24518#issuecomment-4120401246 2026-03-24 11:19:15 -07:00
Krrish DholakiaandGitHub 0af114fb3a Merge pull request #24525 from BerriAI/litellm_pin_trivvy
fix(security_scans.sh): pin trivvy version
2026-03-24 09:15:34 -07:00
Krrish Dholakia fb379865ab fix(security_scans.sh): pin trivvy version 2026-03-24 09:13:37 -07:00
ryan-crabbeandGitHub 764b96b1aa Merge pull request #24486 from BerriAI/litellm_ryan-march-23-audit-log-export
docs: add audit log export to S3 documentation
2026-03-23 22:54:39 -07:00
ryan-crabbeGitHubgreptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
ffbe3896d2 Update docs/my-website/docs/proxy/multiple_admins.md
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
2026-03-23 22:54:08 -07:00
ryan-crabbeandGitHub 4629dfe394 Merge pull request #24485 from BerriAI/litellm_migrate-route-preview-tremor-to-antd
chore: migrate route_preview.tsx from Tremor to Ant Design
2026-03-23 22:51:42 -07:00
Ryan Crabbe 57c027a2dc docs: add audit log export to S3 documentation 2026-03-23 22:46:38 -07:00