Starstruk Logo MarkStarstruk

v0.33.1

SocAIty/SpeechCraftv0.33.1Aug 26, 2026by github-actions[bot]

AI Summary

This release adds Qwen3.8 Flash Next model support and updates MLX and llama.cpp dependencies. It also introduces structured output support to mlxrunner and fixes Metal GPU timeout issues when loading models from slow storage.

Key Highlights

  • MLX: Qwen3.8 Flash Next support
  • mlxrunner: add structured output support
  • mlxrunner: avoid Metal GPU timeouts when loading models from slow storage
  • MLX and llama.cpp update

New Features

  • Qwen3.8 Flash Next support
  • Structured output support in mlxrunner

Full Release Notes

## What's Changed
* MLX: Qwen3.8 Flash Next support
* cmake: make external compat patches idempotent
* MLX and llama.cpp update
* mlxrunner: add structured output support
* mlxrunner: avoid Metal GPU timeouts when loading models from slow storage

## New Contributors
* @pd95 made their first contribution in https://github.com/ollama/ollama/pull/17948

**Full Changelog**: https://github.com/ollama/ollama/compare/v0.33.0...v0.33.1