LiteLLM v1.85.1 Patch Adds Gemini 3.5 Flash Support and Cross-Pod Spend Fix
Below validation threshold — auto-passed without scoring
LiteLLM released v1.85.1 on May 20, 2026, as a patch on top of v1.85.0 with day-0 support for Gemini 3.5 Flash on Google AI Studio and Vertex AI. The update fixed Vertex AI tool-calling HTTP 400 errors for Gemini 3.5+ models and resolved cross-pod spend-counter double-counting after Redis cache misses. The same fixes shipped for the 1.84 line as v1.84.1. Prior v1.85.0 on May 16 added OpenAI Realtime GA, MCP Gateway expansion, and multi-tenancy hardening. LiteLLM continues regular releases supporting 100+ providers with enterprise features.