Instructions to use gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit --local-dir MiniMax-H3-Ref2VA-MLX-Serve-8bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
|
Download MODIFICATIONS.md from gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit: direct link, hf CLI and curl.
- Browser
- Download file 312 Bytes
-
https://huggingface.co/gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit/resolve/main/MODIFICATIONS.md
- Command line
-
hf download hf://gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit/MODIFICATIONS.md
-
curl -L -o MODIFICATIONS.md https://huggingface.co/gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit/resolve/main/MODIFICATIONS.md
312 Bytes
Modifications to MiniMax H3
The REF2VA transformer and text encoder were converted from BF16 to MLX affine 8-bit weights with group size 64. Gathered tables, precision-sensitive projections, output heads and both VAEs remain dense. No weights were retrained or distilled. See LICENSE for the governing terms.