scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-27 01:22:18 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 80
52478 commits 2844 branches 1422 tags 3.6 GiB
Commit graph

13690 commits

Author SHA1 Message Date
mateo
9d201a74eb test(proxy): restore saved_configs capture in team default params test 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 23:33:22 +00:00
mateo
912798c999 Merge remote-tracking branch 'origin/main' into litellm_frozen_proxy_config 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

# Conflicts:
#	tests/test_litellm/proxy/test__types.py
 2026-09-21 23:27:47 +00:00
mateo
108f699294 test(proxy): cover frozen runtime config and section-based save_config 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 23:11:58 +00:00
joshua-berri
1baa26d36b
Merge pull request #41074 from BerriAI/litellm_mcp_server_list_stable_order 
fix(mcp): keep server lists stable across refreshes
 2026-09-21 23:06:09 +00:00
ryan-crabbe-berri
a0d1f0236e
Merge pull request #42122 from BerriAI/litellm_db_callback_redaction_settings 
fix(proxy): apply DB-stored callback redaction settings before logger init
 2026-09-21 16:03:44 -07:00
yujonglee
ced2980529
Merge pull request #42308 from BerriAI/litellm_rust_secrets_hashicorp_vault 
feat(rust): add HashiCorp Vault secret manager crate
 2026-09-21 16:03:33 -07:00
kerry-berri
5fb66e8d98
Merge pull request #42353 from BerriAI/litellm_xai_max_completion_tokens 
fix(xai): accept max_completion_tokens as a supported param
 2026-09-21 15:53:05 -07:00
Yujong Lee
bd3a680469 merge(main): integrate upstream cache dependencies 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 22:51:05 +00:00
ryan-crabbe-berri
0efcb9956d
Merge pull request #42278 from BerriAI/litellm_cherry_pick_password_breach_reset 
feat(auth): breached password detection, self-service change-password and forced password reset
 2026-09-21 15:47:46 -07:00
kerry
fee8f38ccc Merge remote-tracking branch 'origin/main' into litellm_xai_max_completion_tokens 2026-09-21 22:42:12 +00:00
ryan-crabbe-berri
a776d0d722
Merge pull request #41561 from chopratejas/fix-responses-guardrail-custom-tool-output-writeback 
fix(responses): patch custom_tool_call_output in place on guardrail write-back
 2026-09-21 15:40:34 -07:00
kerry
32c63e332e Revert "fix(xai): accept max_completion_tokens as a supported param" 
This reverts commit 4ff251e6b1.
 2026-09-21 22:38:29 +00:00
joshua-berri
3353c57e76
Merge pull request #42292 from BerriAI/litellm_extract_mcp_operation_context_7741 
refactor(mcp): extract explicit operation context and dispatch
 2026-09-21 22:38:07 +00:00
kerry
4ff251e6b1 fix(xai): accept max_completion_tokens as a supported param 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 22:37:40 +00:00
kerry
5c0589207c fix(xai): accept max_completion_tokens as a supported param 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 22:30:13 +00:00
ryan
8a1c6e49fb feat(auth): only allow password-login dashboard sessions to call /user/password/change 
Password login now stamps login_method=username_password into the UI session key
metadata, and change_password rejects any caller that is not a litellm-dashboard
key carrying that marker with 403 before the user row is read. SSO sessions and
user-associated virtual keys can no longer use the endpoint as a current_password
guessing oracle. The forced-reset session is still minted by the password login
path, so it keeps access to the endpoint

Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 22:22:12 +00:00
Mateo Wang
3d26a29a1a
Merge pull request #42283 from BerriAI/litellm_mid_stream_fallback_walks_full_list 
fix(router): walk every entry of a fallback list after a mid-stream failure
 2026-09-21 15:14:04 -07:00
Mateo Wang
7ae61b62ed
Merge pull request #42323 from BerriAI/litellm_stream_usage_keep_explicit_zero 
fix(streaming): keep an explicit provider prompt_tokens=0 or completion_tokens=0 in streamed usage
 2026-09-21 15:13:21 -07:00
ryan-crabbe-berri
f200d68309
Merge pull request #41560 from chopratejas/fix-responses-agentic-followup-duplicate-kwargs 
fix(responses): stop agentic follow-up from passing request params twice
 2026-09-21 15:12:18 -07:00
Mateo Wang
055b7314e0
Merge pull request #40121 from Atharva-Kanherkar/fix/mcp-responses-stream-single-lifecycle 
fix(responses): stream one lifecycle across MCP auto-execute rounds
 2026-09-21 15:11:57 -07:00
Mateo Wang
e7bff277a6
Merge pull request #40251 from BerriAI/litellm_decrease_anys_opus5_r5 
refactor(types): remove 1,173 Any errors across 169 backend files
 2026-09-21 15:11:44 -07:00
Yujong Lee
905909741d merge(main): integrate upstream Azure secret changes 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 22:10:22 +00:00
ryan-crabbe-berri
ddb236b7e2 Merge remote-tracking branch 'origin/main' into pr-41560 
# Conflicts:
#	tests/test_litellm/llms/custom_httpx/test_llm_http_handler.py
 2026-09-21 15:04:57 -07:00
yujonglee
41adbdaa05
Merge pull request #42309 from BerriAI/litellm_rust_secrets_azure_key_vault 
feat(rust): add Azure Key Vault secret manager backend
 2026-09-21 15:04:10 -07:00
ryan
28f70c75d1 fix(auth): reject unchanged password on /user/password/change 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 22:02:05 +00:00
Mateo Wang
7b8bc54237
Merge pull request #42036 from BerriAI/litellm_team_membership_lookup_fail_closed 
fix(auth): fail closed when the team membership lookup hits a db outage
 2026-09-21 14:58:46 -07:00
mateo-berri
e7557ada57 fix(responses): list executed MCP calls as completed mcp_call items in the final output 2026-09-21 14:56:02 -07:00
ryan
7c8aed072f chore: merge main into litellm_cherry_pick_password_breach_reset 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 21:54:35 +00:00
Mateo Wang
4b54897b36
Merge pull request #42315 from BerriAI/litellm_scim_delete_evict_user_cache 
fix(proxy): evict the cached user row when SCIM or /user/delete removes a user
 2026-09-21 14:51:53 -07:00
yucheng-berri
e0e5b7924a
Merge pull request #42312 from BerriAI/litellm_bedrock_batch_s3_creds_leak 
Register s3_access_key_id, s3_secret_access_key and s3_encryption_key_id as
LiteLLM-owned batch params so they are no longer forwarded to Bedrock as
additionalModelRequestFields (which 400s ordinary chat on a batch-configured
deployment), keep them on CredentialLiteLLMParams so the batch/file paths
still receive them, and redact the S3 credential key names in debug logs.

Resolves LIT-8290

Co-authored-by: yucheng <yucheng@berri.ai>
Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 14:50:54 -07:00
Mateo Wang
10d0d5acb2
Merge pull request #42314 from BerriAI/litellm_alerting_batch_keeps_distinct_alerts 
fix(alerting): deliver every distinct alert queued in one flush window
 2026-09-21 14:49:54 -07:00
mateo-berri
b0305d0a31 test(router): call both mid-stream fallback attempt functions directly 
The router coverage gate wants every router.py function reached by name from
a router test. The two per-endpoint attempt functions were only reached through
their callers, so each now has a direct test proving the per-request controls
carrier never reaches the provider call and every hop's stream comes back wrapped.
 2026-09-21 14:48:42 -07:00
Yujong Lee
e0e4850ec1 merge(main): integrate upstream secret manager changes 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 21:48:18 +00:00
Yujong Lee
b926cf998a chore: merge main into litellm_rust_secrets_azure_key_vault 
Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>
 2026-09-21 21:47:21 +00:00
yujonglee
ebb4d230f8
Merge pull request #42300 from BerriAI/litellm_python_interop_foundation 
fix(rust): preserve Python settings semantics at the native boundary
 2026-09-21 14:46:37 -07:00
yujonglee
162e5c4338
Merge pull request #42303 from BerriAI/litellm_rust_secrets_cyberark 
feat(rust): add CyberArk Conjur secret manager backend
 2026-09-21 14:45:39 -07:00
Yassin Kortam
17b56cc4ca
Merge pull request #41505 from BerriAI/litellm_keep_config_models_on_empty_config_read 
fix(proxy): keep config-defined deployments when a config read returns no model_list
 2026-09-21 16:43:15 -05:00
Joshua Valluru
5ec61e7a0b chore: merge main into MCP ordering fix 2026-09-21 14:36:41 -07:00
yucheng-berri
506cecfb0b
Merge pull request #42267 from BerriAI/litellm_otel_v2_langfuse_ocr_output 
fix(otel v2): map OCR page markdown onto the generation output
 2026-09-21 14:33:02 -07:00
Yujong Lee
5aeb367d2a fix(rust): preserve Python settings coercion at the native boundary 2026-09-21 21:32:33 +00:00
yucheng-berri
79d6e236f9
Merge pull request #39805 from BerriAI/litellm_mcp_admin_api_preserve_oauth_scopes 
fix(mcp): keep oauth scopes in admin api credential redaction
 2026-09-21 14:29:13 -07:00
mateo-berri
a2ae80ec9b fix(router): wrap every Responses and Messages fallback hop for mid-stream failover 
The /v1/responses and /v1/messages streaming wrappers only ever wrapped the
primary's stream, so a hop reached through the regular fallback chain had no
mid-stream handler: its failure re-raised, or the outer wrapper retried the
same entry with a fresh attempted set and never reached the rest of the list.
Every attempt of the chain now runs through a per-endpoint attempt function
that wraps its own stream, mirroring chat completions, and the per-request
fallback and retry overrides ride a frozen carrier so each hop's re-entry
still sees them after the retry layer pops them.
 2026-09-21 14:19:52 -07:00
mateo-berri
2f8bee053d test(alerting): inject the webhook client and extend the mapped test files 2026-09-21 14:16:37 -07:00
Yassin Kortam
f6c69af427
Merge pull request #41101 from hMED22/litellm_add_edenai_provider 
feat(edenai): add Eden AI provider across chat, Responses, Messages, embeddings, audio, images and video
 2026-09-21 16:16:28 -05:00
Mateo Wang
fc82f6e8fa
Merge pull request #42288 from BerriAI/litellm_safeguards_bedrock_vertex_messages 
fix(anthropic): forward Claude Code safeguards and dangerous-tool-use beta to Bedrock Invoke and Vertex on /v1/messages
 2026-09-21 14:15:36 -07:00
mateo-berri
1b568319d0 fix(types): blank non-string datadog tool text fields, look the spend table up by name 2026-09-21 14:10:09 -07:00
Yassin Kortam
da1ccaec67
Merge pull request #40322 from BerriAI/litellm_lit7351_reservation_lease_renewal 
fix(proxy): renew budget reservation counter TTL while the request is in flight
 2026-09-21 16:07:14 -05:00
yuneng-jiang
5e0512b611
Merge pull request #42291 from BerriAI/litellm_lit7597_detach_credential 
fix(proxy): detach stored credential when model editor selects None
 2026-09-21 14:05:28 -07:00
mateo-berri
a64febb3e7 fix(streaming): keep an explicit provider prompt_tokens=0 or completion_tokens=0 in streamed usage 
The stream chunk builder started its per-chunk accumulators at 0 and adopted only nonzero counts, then fell back to litellm's tokenizer whenever the accumulated value was falsy, so a provider that reported an explicit 0 for prompt or completion tokens was billed the estimate instead. The accumulators now start at None, a usage chunk that reports a count marks it reported (a later chunk's 0 never replaces a reported nonzero), and the estimate only runs when no chunk reported the count. The Anthropic message_start cursor reset now yields None so the estimate still covers a cancelled stream, and Ollama chat streaming only attaches usage on the done chunk when both counts are present instead of inventing 0/0 on every chunk
 2026-09-21 14:01:08 -07:00
kerry-berri
89de508d6e
Merge pull request #42306 from BerriAI/litellm_fal_ai_surface_video_errors 
fix(fal_ai): surface fal errors in video status and content instead of completed and generic 500
 2026-09-21 13:59:04 -07:00