Video-Text-to-Text
MLX
Safetensors
English
qwen3_5
video
multimodal
video-captioning
temporal-grounding
quantized
8-bit precision
custom_code
8-bit precision
Instructions to use NemoStation/Marlin-2B-MLX-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use NemoStation/Marlin-2B-MLX-8bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir Marlin-2B-MLX-8bit NemoStation/Marlin-2B-MLX-8bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
fix(processor): Qwen3VL*->Qwen2VL* image/video processor classes (match base + preprocessor_config; fixes AutoProcessor on sglang-mlx)
1434587 verified | { | |
| "image_processor": { | |
| "do_convert_rgb": true, | |
| "do_normalize": true, | |
| "do_rescale": true, | |
| "image_mean": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "image_processor_type": "Qwen2VLImageProcessor", | |
| "image_std": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "max_pixels": 16777216, | |
| "merge_size": 2, | |
| "min_pixels": 65536, | |
| "patch_size": 16, | |
| "rescale_factor": 0.00392156862745098, | |
| "temporal_patch_size": 2 | |
| }, | |
| "processor_class": "Qwen3VLProcessor", | |
| "video_processor": { | |
| "do_convert_rgb": true, | |
| "do_normalize": true, | |
| "do_rescale": true, | |
| "fps": 2.0, | |
| "image_mean": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "image_std": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "max_frames": 768, | |
| "max_pixels": 16777216, | |
| "merge_size": 2, | |
| "min_frames": 4, | |
| "min_pixels": 65536, | |
| "patch_size": 16, | |
| "rescale_factor": 0.00392156862745098, | |
| "temporal_patch_size": 2, | |
| "video_processor_type": "Qwen2VLVideoProcessor" | |
| } | |
| } |