Local AI Models

Run AI models locally — Ollama, LM Studio, vLLM

6 Updates

feature

Ollama v0.32.4 adds MLX Apple GPU support for Laguna and optimizes Qwen3 MoE decoding.

27 July 2026Source

feature

LM Studio 0.4.20 adds support for enterprise internal network endpoints and LM Link model sharing with Bionic.

27 July 2026Source

feature

LocalAI v4.7.1 introduces UI-managed voice cloning profiles, local avatar generation, and improved GPU device controls.

27 July 2026Source

feature

Ollama v0.32.0 launches an interactive agent experience directly in the CLI for chatting, coding, and web searching.

27 July 2026Source

feature

LM Studio 0.4.20 introduces support for enterprise internal network model endpoints and using local models in Bionic over LM Link.

27 July 2026Source

feature

LocalAI v4.7.0 introduces a UI-managed voice cloning library, local avatar generation, and interleaved thinking with tool calls.

27 July 2026Source