PipeNetwork/minimax-h3-mlx
PipeNetwork/minimax-h3-mlx MiniMax released MiniMax-H3 two days ago - they describe it as a "a general-purpose, omni-modal generative system", which in practice means it accepts text, images, audio and video and can use them to generate up to 15 second video clips with audio included. This Python package ports it to MLX for running on Apple Silicon. I got it running on my M5 Max MacBook Pro. I…
MiniMax introduced MiniMax-H3 on August 4th, 2026, describing it as a versatile generative system capable of processing text, images, audio, and video to produce up to 15-second video clips with incorporated audio. This Python package adapts MiniMax-H3 to MLX, enabling it to run on Apple Silicon devices. The author successfully ran the model on their M5 Max MacBook Pro, cloning the repository and initiating the model with a prompt: "a rainbow colored skunk leaps over a mossy log in a supermarket."
The download of approximately 115 GB of model files took some time, and the subsequent video generation took just under 45 minutes. While the video output was impressive, the accompanying audio was a garbled speech-like noise due to the lack of specific audio guidance in the prompt. The author did not read the provided prompting guide before this experiment, but it contains detailed instructions for optimizing the model's performance.
Written by urgent.news from Simon Willison's reporting — not their text. Machine-written — it may contain errors, so check the original before relying on it.
