Image-Text-to-Text
MLX
Safetensors
nemotron_parse
ocr
document-parsing
vision-encoder-decoder
multimodal
custom_code
8-bit precision
Instructions to use mlx-community/Nemotron-Parse-2.0-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/Nemotron-Parse-2.0-8bit with MLX:
# Make sure mlx-vlm is installed # pip install --upgrade mlx-vlm from mlx_vlm import load, generate from mlx_vlm.prompt_utils import apply_chat_template from mlx_vlm.utils import load_config # Load the model model, processor = load("mlx-community/Nemotron-Parse-2.0-8bit") config = load_config("mlx-community/Nemotron-Parse-2.0-8bit") # Prepare input image = ["http://images.cocodataset.org/val2017/000000039769.jpg"] prompt = "Describe this image." # Apply chat template formatted_prompt = apply_chat_template( processor, config, prompt, num_images=1 ) # Generate output output = generate(model, processor, formatted_prompt, image) print(output) - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
| { | |
| "format": "safetensors", | |
| "description": "Auxiliary decoder prediction-head tensors retained from training. These are not loaded by the default HF/vLLM inference path.", | |
| "tensor_file": "auxiliary_prediction_heads.safetensors.extra", | |
| "tensors": { | |
| "decoder.extra_heads.0.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "dtype": "float32" | |
| }, | |
| "decoder.extra_heads.0.weight": { | |
| "shape": [ | |
| 1024, | |
| 1024 | |
| ], | |
| "dtype": "float32" | |
| }, | |
| "decoder.extra_proj.0.bias": { | |
| "shape": [ | |
| 1024 | |
| ], | |
| "dtype": "float32" | |
| }, | |
| "decoder.extra_proj.0.weight": { | |
| "shape": [ | |
| 1024, | |
| 1024 | |
| ], | |
| "dtype": "float32" | |
| } | |
| } | |
| } | |