§ tools · storyline

Adds Laguna on Apple GPUs via MLX engine

Ollama adds Laguna support on Apple GPUs via the MLX engine.

yesterday · 04:22:12 · primary fetch1 sourceupdated yesterday · 04:22:12

What's Changed Support Laguna on Apple GPUs via the MLX engine Quantize draft-model output heads at the requested type when creating speculative-decoding drafts. Fixed Qwen3 MoE decoding for differently-quantized experts, plus faster packed gate/up projection (~4–9% on M5 Max).

Full Changelog: https://github.com/ollama/ollama/compare/v0.32.3...v0.32.4

read full article on github.com ↗

§ sources1 publication · timeline below

github.comOllama v0.32.4primary04:22:12