-
-
Notifications
You must be signed in to change notification settings - Fork 7.4k
Pull requests: BerriAI/litellm
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
fix(logging): create spend logs for native /v1/messages handler
#26114
opened Apr 20, 2026 by
Vigilans
Loading…
3 of 4 tasks
fix(adapter): normalize reasoning effort with graceful degradation
#26111
opened Apr 20, 2026 by
Vigilans
Loading…
3 of 4 tasks
fix: post call guardrail must be called once
#26109
opened Apr 20, 2026 by
mubashir1osmani
Contributor
Loading…
7 tasks
fix: drop orphan tool_choice after web_search conversion & bridge web_search_options to Responses API where appropriate
#26107
opened Apr 20, 2026 by
Vigilans
Loading…
3 of 4 tasks
fix(langfuse): send v4 ingestion header for otel callback
#26105
opened Apr 20, 2026 by
hassiebp
Loading…
fix(responses): emit per-summary thinking blocks and deduplicate message_start
#26103
opened Apr 20, 2026 by
Vigilans
Loading…
3 of 4 tasks
fix(anthropic): allow max output effort via model metadata
#26101
opened Apr 20, 2026 by
Sameerlite
Collaborator
Loading…
2 tasks done
fix(vertex-ai): reuse anthropic messages config instances
#26099
opened Apr 20, 2026 by
Sameerlite
Collaborator
Loading…
1 task done
fix(ollama): process both thinking and content in same streaming chunk
#26098
opened Apr 20, 2026 by
vhash0
Loading…
feat: add
force_reasoning_effort litellm_params config
#26096
opened Apr 20, 2026 by
Vigilans
Loading…
4 tasks done
fix(mcp): resolve client-prefixed MCP tool names in semantic filter
#26095
opened Apr 20, 2026 by
XyLearningProgramming
Loading…
5 of 7 tasks
docs: add Claude Code any-advisor tutorial and fix streaming logging
#26092
opened Apr 20, 2026 by
Sameerlite
Collaborator
Loading…
3 tasks done
fix(anthropic): preserve tool blocks in messages streaming
#26090
opened Apr 20, 2026 by
Rookie0x80
Loading…
1 task done
fix(proxy): paginate reset_budget_windows scans
#26089
opened Apr 20, 2026 by
jibanez-staticduo
Loading…
fix(messages): route to Responses API based on model_info.mode
#26088
opened Apr 20, 2026 by
Vigilans
Loading…
4 tasks done
fix(responses): handle incomplete response with reasoning-only output
#26087
opened Apr 20, 2026 by
Vigilans
Loading…
4 tasks done
fix(messages): copy metadata dict to prevent raw_request leak to Anthropic API
#26086
opened Apr 20, 2026 by
Vigilans
Loading…
4 tasks done
fix(proxy): avoid invalid Prisma JsonFilter in reset_budget_windows
#26085
opened Apr 20, 2026 by
jibanez-staticduo
Loading…
fix(model-registry): add missing supports_web_search for gpt-5.4
#26084
opened Apr 20, 2026 by
Vigilans
Loading…
4 tasks done
fix(pass-through): pop stale route key directly to prevent unbounded registry growth
#26082
opened Apr 20, 2026 by
silencedoctor
Contributor
Loading…
3 tasks done
feat(proxy): add model_group filter to /spend/logs/v2 endpoint
#26080
opened Apr 20, 2026 by
silencedoctor
Contributor
Loading…
3 tasks done
fix(vertex_ai): omit system_instruction/tools/toolConfig when cachedContent set
#26077
opened Apr 20, 2026 by
Sameerlite
Collaborator
Loading…
feat(vertex_ai): Model Garden OpenAPI for publisher model ids
#26076
opened Apr 20, 2026 by
Sameerlite
Collaborator
Loading…
fix(chatgpt): accumulate response.output_item.done for non-streaming responses
#26075
opened Apr 20, 2026 by
andafterall
Loading…
5 tasks done
Previous Next
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.