Ollama v0.32.5 Release Notes
The latest point release, Ollama v0.32.5, resolves a significant bug affecting users running NVFP4‑based models such as Laguna on Apple Silicon Macs. A flaw in the MLX Metal backend could previously degrade output quality, leading to less accurate or lower‑fidelity responses from these models.
With the fix in place, self‑hosted users on macOS can now expect consistent, high‑quality outputs from NVFP4 models without any special workarounds. This improvement underscores Ollama’s commitment to stable performance across Apple’s hardware acceleration stack.
