LightNews — News, ranked by scholars

Simon Willison

@simonwillison.net

New LLM plugin for running models on macOS using Apple's MLX framework:

llm install llm-mlx
llm mlx download-model mlx-community/Llama-3.2-3B-Instruct-4bit
llm -m mlx-community/Llama-3.2-3B-Instruct-4bit 'Python code to traverse a tree'

simonwillison.net/2025/Feb/15/...

Run LLMs on macOS using llm-mlx and Apple’s MLX framework

llm-mlx is a brand new plugin for my LLM Python Library and CLI utility which builds on top of Apple’s excellent MLX array framework library and mlx-lm package. If you’re …

simonwillison.net

February 15, 2025 at 7:14 PM

Sung Kim

@sungkim.bsky.social

This seems to be the most popular alternative to jev.

Laya: Multilingual, non-autoregressive System 1 decision model.
huggingface.co/convaiinnova...
Laya Repo: github.com/NandhaKishor...

Laya MLX: huggingface.co/aac6fef/laya...
Laya MLX Repo: github.com/mizorewww/la...

September 22, 2026 at 5:29 AM

noob 🌈

@noobfl.bsky.social

hier, dingens

bsky.app/profile/mlx....

mlx @mlx.bsky.social · Sep 16

dennis krapp wohnt jetzt in berlin. spannend.

"Böses Omen

Die Einnahmen aus Techno-Partys zu Ehren des legendären Omen sollten gespendet werden. Doch bei den karitativen Organisationen kam nie Geld an."

September 19, 2026 at 6:10 AM

amos

@fasterthanli.me

SpeechLens looks really cool (MLX adaption of RE-USE), tl;dr removes background noise from voice recordings. I don't really have a use for it yet because... of my really good audio recording set up. But. It's really neat.

github.com/faraday/Spee...

GitHub - faraday/SpeechLens: Fast on-device speech enhancement for Mac, powered by RE-USE, MLX & Metal

Fast on-device speech enhancement for Mac, powered by RE-USE, MLX & Metal - faraday/SpeechLens

github.com

September 24, 2026 at 12:53 PM

omnia pro omnibus

@ravachol-161.bsky.social

Hat scheinbar Tradition:

bsky.app/profile/mlx....

mlx @mlx.bsky.social · Sep 16

dennis krapp wohnt jetzt in berlin. spannend.

September 19, 2026 at 7:24 AM

mr. TIM

@timkellogg.me

mlx-knife: an ollama-like CLI for Apple Silicon

alright, this is the end of the road for me & ollama

github.com/mzau/mlx-knife

GitHub - mzau/mlx-knife: ollama like cli tool for MLX models on huggingface (pull, rm, list, show, serve etc.)

ollama like cli tool for MLX models on huggingface (pull, rm, list, show, serve etc.) - mzau/mlx-knife

github.com

September 1, 2025 at 11:54 AM

Simon Willison

@simon.fedi.simonwillison.net.ap.brid.gy

New LLM plugin for running models on macOS using Apple's MLX framework:

llm install llm-mlx
llm mlx download-model mlx-community/Llama-3.2-3B-Instruct-4bit
llm -m mlx-community/Llama-3.2-3B-Instruct-4bit 'Python code to traverse a tree'

https://simonwillison.net/2025/Feb/15/llm-mlx/

Run LLMs on macOS using llm-mlx and Apple’s MLX framework

llm-mlx is a brand new plugin for my LLM Python Library and CLI utility which builds on top of Apple’s excellent MLX array framework library and mlx-lm package. If you’re …

simonwillison.net

February 15, 2025 at 7:45 PM

MLX-Junko

@junkomlx.bsky.social

I'm MLX-Junko. I draw art of cute and sexy girls~
#portfolioday

October 15, 2025 at 3:52 AM

Simone Conradi

@sconradi.bsky.social

3D embedding of my polynomial roots #mathart.
Made with #python #apple #MLX and #pyGFX.

July 8, 2025 at 7:20 AM

Simone Conradi

@sconradi.bsky.social

Particle Lenia simulation.
My code: github.com/profConradi/...
Made with #python, #matplotlib and #mlx.

November 21, 2024 at 12:14 PM

Joshua White

@jrw14.whnc.me

Gosh, I really love the opensource community.

This morning I was thinking to myself, if my potato M1 Mac Mini was worth keeping on for just an embedding model and ling-tiny-3.0 LLM (for very light background tasks).

Then I found github.com/mizorewww/la.... Faster and soveriegn Jev?

Yes, please.

GitHub - mizorewww/laya-mlx: Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API.

Native MLX runtime for Laya typed decision models — 7–14 ms short decisions on M3 Max. No text generation, PyTorch, or cloud API. - mizorewww/laya-mlx

github.com

September 22, 2026 at 4:54 PM