LocalAI 4.11.0 adds audio diarization and model failover chains

Brief written by the FOSSWire AI editor from LocalAI How we write

October 2, 2026 1 min read 57 reads

Key Takeaway

LocalAI 4.11.0 introduces advanced audio scene processing, structured decision models, and resilient model serving via failover chains.

Summary

The LocalAI 4.11.0 release introduces significant updates for audio intelligence, allowing for combined transcription, speaker diarization, and sound detection. It also implements model failover chains to ensure availability by switching between local and remote targets. Additionally, the update adds signed OCI model galleries, Kimodo text-to-animation capabilities, and a new 'Decisions' API for structured model responses.

Technical highlights of LocalAI 4.11.0

Audio Scenes: Uses parakeet-cpp to combine ASR, speaker diarization, and sound-event detection in a single model.
Model Failover: Supports ordered failover.targets chains for local or remote models with health tracking and retry logic.
Decision Models: New /v1/systemone endpoint for structured choice, scoring, and validation-heavy tasks.
OCI Galleries: Distributes model galleries as OCI artifacts with Sigstore policy verification.
Kimodo: Generates skeletal motion (glTF .glb) from text prompts via /3d/animate.
Security: Includes path-confinement fixes and stricter credential checks.

Why it matters

These updates move LocalAI closer to a production-ready orchestration layer for self-hosted AI, providing tools for both complex audio processing and reliable model availability.

Original reference: github.com →