LiteLLM 1.108.12 released
An LTS patch release fixing prompt caching, model routing, batch processing, and tool handling across multiple providers.
What changed in 1.108.12
- OpenAI GPT-6.2 prompt caching now handles system prompts correctly with store: false parameter
- Azure AI model router fixed for Foundry endpoints; Vertex AI batch processing corrected for JSON schema rows
- Messages API tool reference errors resolved for names over 64 characters; upstream streaming no longer continues after caller closes connection
- Guardrails tool policy now correctly blocks tools with names exceeding 64 characters
Summarised by AgentGid from BerriAI's official release notes on GitHub, as reported there; read the original before relying on it. Price, status and release dates come from our daily data.
LiteLLM at a glance
| What it is | An AI gateway and Python SDK that calls 100+ model APIs through one OpenAI-style interface, with spend tracking, budgets, rate limits, fallbacks, guardrails and logging. The same gateway also serves A2A agents and MCP t… |
|---|---|
| Latest version | 1.108.14, Nov 20, 2026. This page is about an earlier release. |
| Release pace | 44 stable releases in the last 30 days; 1.108.12 came the same day as the previous one |
| Price | Free (OSS). The SDK and proxy are open source and free. An enterprise edition with SSO, audit logs and support is sold separately. Official pricing ↗ (checked Nov 21, 2026) |
| Downloads, last 7 days | 25.2M up 5.5% |
| Gid Score | 74 #1 in Agent tooling for cost, context and memory |