aiqbench takeAI Products
aiqbench analysis
The update is mainly a release recap for LiteLLM, an LLM gateway, and its most AI-specific item is the new `/v1/decisions` endpoint, which returns per-answer probabilities from any configured model. Readers running LiteLLM should note the compromised `litellm==1.107.0` release and the two opt-in flags that must be set to re-enable STDIO MCP servers and `/sso/debug`. Most other figures, including the bug-fix count, coverage, and test totals, are self-reported and unverified.
1
Source coveringLiteLLM Blog
October Townhall Updates: 426 Bug Fixes, 88.4% Coverage, and the Decisions API
LiteLLM Blog3 min read
A recap of the October LiteLLM town hall: security changes that need action before you upgrade, 426 bug fixes, test coverage at 88.4%, the Decisions API, Lens, Moyai, and LiteLLM Relay.
Thank you to everyone who joined our October town hall. We covered security, stability and performance work, and the product updates from the month. On October 21 our release pipeline published a compromised litellm==1.107.0. The attack chain post has the full record and what to do. Three changes need action before you upgrade. STDIO MCP servers are off. LITELLM_ENABLE_MCP_STDIO=true turns them on /sso/debug is…
Excerpt shown under the publisher's RSS terms · 64 words · full text and figures are on the publisher's site