scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-08 03:08:45 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 115
39084 commits 3285 branches 1443 tags 4.8 GiB
Commit graph

39084 commits

This branch
This branch
All branches
Author SHA1 Message Date
yuneng-jiang
c89496f378
Merge pull request #24342 from BerriAI/litellm_yj_march_21_2026 
[Fix] UI - Teams: Table refresh, infinite dropdown, leftnav migration
 2026-03-21 23:36:21 -07:00
yuneng-jiang
38d477507d remove outdated e2e test 2026-03-21 23:14:53 -07:00
yuneng-jiang
6bb08883af adding poetry lock 2026-03-21 23:01:28 -07:00
yuneng-jiang
fa65433c8c bump: version 1.82.5 → 1.82.6 2026-03-21 22:56:09 -07:00
yuneng-jiang
d91980dc45 adding build 2026-03-21 22:55:04 -07:00
yuneng-jiang
071c8641de bump: version 0.4.59 → 0.4.60 2026-03-21 22:54:41 -07:00
yuneng-jiang
88a4c7aeaf bump: version 0.4.58 → 0.4.59 2026-03-21 22:54:38 -07:00
yuneng-jiang
34d079910b chore: update Next.js build artifacts (2026-03-22 05:53 UTC, node v22.16.0) 2026-03-21 22:53:50 -07:00
yuneng-jiang
9073daeebc [Fix] UI - TeamDropdown: Match org dropdown styling and fix test mock 
- Use Select.Option with font-medium alias + Text secondary ID to match OrganizationDropdown
- Default page size to 20
- Add useInfiniteTeams mock to AddModelForm tests

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-21 22:52:16 -07:00
yuneng-jiang
aea8e32048 [Fix] UI: Team table refresh, infinite team dropdown, leftnav for dashboard routes 
- OldTeams: refresh table via fetchTeamsV2 after team create instead of appending
- TeamDropdown: rewrite with useInfiniteTeams for paginated fetch, scroll-to-load, and debounced search
- Update all TeamDropdown consumers to use the new self-fetching API
- Dashboard layout: switch from Sidebar2 to SidebarProvider (leftnav)
- Leftnav: add MIGRATED_PAGES routing for path-based navigation (api-reference)
- Navbar: remove chat button

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-21 22:05:26 -07:00
Chesars
b038c0714c refactor: extract _flatten_and_detect_file_refs helper, warn on ModuleNotFoundError 2026-03-22 01:52:44 -03:00
Chesars
7091f87a7a fix: rename test method 2026-03-22 01:37:59 -03:00
yuneng-jiang
9963b31e07 Revert "fix(proxy): restore per-entity breakdown in aggregated daily activity endpoint" 
This reverts commit 9c3fab24ad.
 2026-03-21 21:37:29 -07:00
Chesars
11e4ec00f7 fix: resolve file refs in nested lists, only flag truly multimodal nested inputs 2026-03-22 01:37:18 -03:00
Cesar Garcia
912f08b61d
Merge pull request #24341 from Chesars/feat/gemini-combined-multimodal-embeddings 
feat(gemini): support combined multimodal embeddings via nested input
 2026-03-22 01:13:28 -03:00
Chesars
9fc6fdad66 fix: clear error for nested lists on embedContent path, add validation tests 2026-03-22 01:08:01 -03:00
Chesars
960108939d fix: update EmbeddingInput type, validate nested sub-elements, add tests 2026-03-22 01:01:00 -03:00
Chesars
ca37ced620 feat(gemini): support combined multimodal embeddings via nested input 
Allows wrapping multiple inputs in a nested list to produce a single
combined embedding (text + image = 1 vector). Flat lists continue to
produce separate embeddings per input (OpenAI-compatible default).

Examples:
  input=["text", "image"]        → 2 separate embeddings
  input=[["text", "image"]]      → 1 combined embedding
  input=[["text", "image"], "x"] → 2 embeddings (1 combined + 1 separate)
 2026-03-22 00:53:51 -03:00
Chesars
4694a30b62 refactor: remove unnecessary sys.path.insert from test file 2026-03-22 00:27:39 -03:00
Chesars
76d53fe861 fix: raise clear error when file references used without api_key 2026-03-22 00:24:38 -03:00
Chesars
c91b44adeb fix: ensure gitignored model_prices JSON is included in wheel/sdist 2026-03-22 00:14:29 -03:00
Cesar Garcia
62df6325b1
Merge pull request #16590 from Chesars/refactor/remove-backup-file-dry-principle 
refactor: Remove redundant backup file
 2026-03-21 23:57:58 -03:00
Chesars
556a1ae144 Merge remote-tracking branch 'upstream/real-main' into refactor/remove-backup-file-dry-principle 
# Conflicts:
#	litellm/model_prices_and_context_window_backup.json
 2026-03-21 23:52:07 -03:00
Chesars
965b952599 refactor: rename 'backup' to 'local' in get_model_cost_map, align test loader 2026-03-21 23:49:19 -03:00
Cesar Garcia
a6462143be
Merge pull request #24337 from Chesars/fix/gemini-multimodal-batch-embeddings-24209 
fix(gemini): return separate embeddings for multimodal inputs
 2026-03-21 23:42:21 -03:00
Chesars
883e150804 fix: count text tokens only for mixed multimodal inputs, add embed_content tests 2026-03-21 23:37:34 -03:00
Chesars
bb247685da fix: skip token counting for multimodal inputs in process_response 2026-03-21 23:29:45 -03:00
Cesar Garcia
74016362bd
Merge pull request #24338 from Chesars/feat/bedrock-glm-5-24227 
feat(bedrock): add Z.AI GLM-5 model support
 2026-03-21 23:27:01 -03:00
Chesars
9c4c34177e refactor: reuse _build_part_for_input in embed_content transform 2026-03-21 23:23:41 -03:00
Chesars
806dd31158 test: add multimodal mixed input test for process_response 2026-03-21 23:19:20 -03:00
Chesars
a6ef217728 feat(bedrock): add Z.AI GLM-5 model support (#24227) 
Add zai.glm-5 to model_prices_and_context_window.json with Bedrock
Converse routing. Pricing: $1.00/1M input, $3.20/1M output tokens.
 2026-03-21 23:11:03 -03:00
Chesars
410f54dc72 fix(gemini): return separate embeddings for multimodal batch inputs (#24209) 
When multiple inputs were passed to the Gemini embedding endpoint and any
contained multimodal data (images, audio, etc.), LiteLLM incorrectly used
the `embedContent` endpoint which combines all inputs into a single
aggregated embedding. Now uses `batchEmbedContents` with each input as a
separate request, returning N embeddings for N inputs as expected.

Also fixes hardcoded index=0 in batch embedding responses.
 2026-03-21 23:06:24 -03:00
yuneng-jiang
f5194b5ce3
Merge pull request #24334 from BerriAI/litellm_yj_march_21_2026 
[Infra] Build UI for release
 2026-03-21 18:03:38 -07:00
yuneng-jiang
d217e49193 chore: update Next.js build artifacts (2026-03-22 01:02 UTC, node v22.16.0) 2026-03-21 18:02:32 -07:00
Cesar Garcia
6cd3896fa6
Merge pull request #24282 from kimimgo/fix/openrouter-prefix-strip-24234 
fix: strip 'openrouter/' prefix from model names (#24234)
 2026-03-21 21:39:33 -03:00
yuneng-jiang
e3d4c29d37
Merge pull request #24323 from BerriAI/litellm_ryan_march_20 
litellm ryan march 20
 2026-03-21 15:57:28 -07:00
Ryan Crabbe
3e27ff1b78 fix: resolve mypy type errors in audit_logs.py 2026-03-21 15:42:01 -07:00
Ryan Crabbe
f494ab513f docs: add High Availability Control Plane documentation 
New docs page covering the HA control plane architecture where each
worker instance has its own DB, Redis, and master key. Includes a
React component diagram, setup configs, SSO notes, and local testing
instructions.
 2026-03-21 15:31:49 -07:00
yuneng-jiang
1986f1034e
Merge pull request #24211 from BerriAI/litellm_dev_sameer_16_march_week 
Litellm dev sameer 16 march week
 2026-03-21 15:19:15 -07:00
yuneng-jiang
72fba093c8 Merge remote-tracking branch 'origin/main' into litellm_dev_sameer_16_march_week 2026-03-21 15:11:29 -07:00
yuneng-jiang
7b31ea40a9
Merge pull request #22844 from BerriAI/litellm_oss_staging_03_05_2026 
Litellm oss staging 03 05 2026
 2026-03-21 15:09:48 -07:00
Ryan Crabbe
9b90e80f71 fix: resolve mypy type errors in audit_logs.py 
Extract multiline `or` chain from LiteLLM_AuditLogs constructor to fix
pydantic mypy plugin field-type misattribution, and add explicit
Optional[bool] annotation to avoid variable name shadowing conflict.
 2026-03-21 15:08:47 -07:00
yuneng-jiang
c6bb7a5be5 Apply Black formatting to 11 files 
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-21 15:05:53 -07:00
yuneng-jiang
e3b62c0915 fix: apply Black formatting to 6 files after main merge 
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-21 15:03:06 -07:00
yuneng-jiang
2b889f1627
Merge pull request #23471 from michelligabriele/fix/aggregated-activity-entity-breakdown 
fix(proxy): restore per-entity breakdown in aggregated daily activity endpoint
 2026-03-21 14:59:41 -07:00
yuneng-jiang
10b0139bf8
Merge branch 'main' into litellm_oss_staging_03_05_2026 2026-03-21 14:58:11 -07:00
Krish Dholakia
f911d8d865
Merge pull request #23818 from BerriAI/litellm_oss_staging_03_17_2026 
fix(fireworks): skip #transform=inline for base64 data URLs (#23729)
 2026-03-21 14:54:39 -07:00
Krrish Dholakia
cb4027531b fix: add explicit "summary" not in result guards to opt-out test paths 
Addresses Greptile feedback that test assertions were weakened when
removing summary: "detailed" expectations — now every default-behavior
test explicitly asserts that "summary" is absent from the result.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-21 14:53:25 -07:00
yuneng-jiang
35316e115f fix: apply Black formatting to 7 files 
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
 2026-03-21 14:51:15 -07:00
yuneng-jiang
262534a3a5
Merge branch 'main' into litellm_dev_sameer_16_march_week 2026-03-21 14:30:57 -07:00