litellm
1117 issues found
[Feature]: Add native DashScope / Alibaba Cloud image generation support for Qwen-Image and Wan models
5/25/2026
[Bug]: Semantic MCP Tool Filtering emits invalid OpenAI tool schema
5/25/2026
RFC: Cross-provider token-budget normalization at the Router layer (Router.budget + Router.check_fit preflight)
5/25/2026
Feature Request: Add Chinese (Simplified) language support / 添加简体中文语言支持
5/25/2026
[Bug]: The integration with RAGFlow isn't working
5/25/2026
[Bug]: Team doesn't exist in db. Team=litellm-dashboard
5/25/2026
[Bug]: streamGenerateContent passthrough always returns SSE format, breaking google-genai SDK
5/25/2026
[Bug]: When an agent with tools turns on either redis-semanticor qdrant-semantic cache mode and carries tool return JSON requests to the gateway, the tool return content will be lost.
5/25/2026
HIGH: Hardcoded DB password dbpassword9090 in docker-compose - stores all LLM API keys
5/25/2026
LOW: Missing SECURITY.md - AI gateway handling LLM API keys
5/25/2026
fix(openai_gpt_5): support custom deployment names with prefixes in is_model_gpt_5_4_plus_model
5/25/2026
CRITICAL: Master API key defaults to sk-1234 — full gateway compromise to 100+ LLM providers
5/25/2026
HIGH: Database password hardcoded as dbpassword9090 in docker-compose.yml
5/25/2026
HIGH: No startup validation prevents running with default master key sk-1234 in production
5/25/2026
CRITICAL (CVSS 9.8): Master API key defaults to sk-1234 — full AI Gateway compromise
5/25/2026
Conflict with **K8s Gateway API Inference Extension**
5/26/2026
[Bug]: Docker image pins glibc version in /etc/apk/world, preventing security upgrades
5/24/2026
[Bug]: Synthetic include_usage chunk violates OpenAI spec — usage event has non-empty choices instead of choices: []`
5/24/2026
[Bug]: /eu.assemblyai passthrough resolves to default US AssemblyAI endpoint
5/24/2026
[Feature]: Project-scoped end-user/customer budget limits
5/24/2026