search

BerriAI / litellm

Python SDK, Proxy Server (LLM Gateway) to call 100+ LLM APIs in OpenAI format - [Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic, Sagemaker, HuggingFace, Replicate, Groq]
https://docs.litellm.ai/docs/
Other
14.08k stars 1.66k forks source link

fix: duplicate exception_type for gemini #6768

Open nobu007 opened 5 days ago

nobu007 commented 5 days ago

Type

🐛 Bug Fix 🧹 Refactoring

Changes

The exception_type for "gemini" is handled same as "palm" now. Original code of "gemini" was in both "vertex_ai" and "palm".

Recentry, gemini often throws 503, so adding switch case as vertex_ai.

This is sample output. Note: logger_fn of exception_logging() is always None. Understanding was not easy for me. ‘‘‘ {'exception': litellm.BadRequestError: GeminiException - { "error": { "code": 503, "message": "The model is overloaded. Please try again later.", "status": "UNAVAILABLE" } } ‘‘‘

vercel[bot] commented 5 days ago

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name Status Preview Comments Updated (UTC)
litellm ✅ Ready (Inspect) Visit Preview 💬 Add feedback Nov 16, 2024 3:50pm