matt-k-wong / mlx-flash
Flash weight streaming for MLX: run massive models larger than your RAM on Apple Silicon.
Why this job: topics: apple-silicon, llm-inference, mlx
What it does
Unlike previous iterations that attempted to re-implement the transformer loop manually, mlx-flash now uses a Holistic Model Patching architecture.
Get started
pip install git+https://github.com/matt-k-wong/mlx-flash.git - Maintenance
- Active
- Last commit 5 months ago
- Stars
- 142
- +1 this week
- Licence
- MIT
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.4.0
- Released 5 months ago
- Built with
- Python
- 94% of the code
Is it looked after?
- Contributors
- 0
- Last 90 days
- Issues
- None
- No issues filed
- Pull requests
- 0 merged
- 0 waiting
- Releases
- 2
- 2 in the last year
- Contributing guide (opens in new tab)
- Security policy (opens in new tab)
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| vllm-project / vllm-metal | 1,979 | yesterday | Apache-2.0 | Mac |
| incoai / splash | 2,170 | 4 days ago | Apache-2.0 | Mac |
| jundot / omlx | 23,491 | 5 days ago | Apache-2.0 | Mac |
More: languages, topics and project record
Languages
- Python 94%
- Metal 4%
- Shell 2%
Topics
- apple-silicon
- large-language-models
- llm
- llm-inference
- lm-studio
- machine-learning
- macos
- memory-optimization
- metal
- mlx
- optimization
- weight-streaming
Project record
- JobLocal LLMs
- StatusListed
- Commits104
- Forks11
- Watching1
- Stars in 28 days+4
- CreatedMar 20, 2026
- First seen by usNov 7, 2026
- CheckedNov 7, 2026