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] huggingface-cli download --local-dir MiniMax-H3-Ref2VA-MLX-Serve-8bit gabrielrocco/MiniMax-H3-Ref2VA-MLX-Serve-8bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
| { | |
| "model_type": "minimax_h3", | |
| "partition": "ref2va", | |
| "tasks": [ | |
| "ref2va" | |
| ], | |
| "sigma_shift_scales": { | |
| "video": 12.0, | |
| "audio": 3.0 | |
| }, | |
| "fps": 24, | |
| "quantization": { | |
| "group_size": 64, | |
| "bits": 8, | |
| "mode": "affine" | |
| }, | |
| "transformer": { | |
| "hidden_size": 5376, | |
| "num_layers": 50, | |
| "num_attention_heads": 56, | |
| "attention_head_dim": 128, | |
| "ffn_hidden_size": 14336, | |
| "latents_dim": 24, | |
| "audio_latents_dim": 32, | |
| "text_dim": 5120, | |
| "time_embed_dim": 2688, | |
| "rope_inv_freq_len": 16 | |
| }, | |
| "text_encoder": { | |
| "hidden": 5120, | |
| "layers": 50, | |
| "heads": 64, | |
| "kv_heads": 8, | |
| "head_dim": 128, | |
| "intermediate": 25600, | |
| "theta": 5000000.0 | |
| } | |
| } |