Instructions to use avbiswas/sam2.1-hiera-tiny-mlx-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use avbiswas/sam2.1-hiera-tiny-mlx-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir sam2.1-hiera-tiny-mlx-4bit avbiswas/sam2.1-hiera-tiny-mlx-4bit
- sam2
How to use avbiswas/sam2.1-hiera-tiny-mlx-4bit with sam2:
# Use SAM2 with images import torch from sam2.sam2_image_predictor import SAM2ImagePredictor predictor = SAM2ImagePredictor.from_pretrained(avbiswas/sam2.1-hiera-tiny-mlx-4bit) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): predictor.set_image(<your_image>) masks, _, _ = predictor.predict(<input_prompts>)# Use SAM2 with videos import torch from sam2.sam2_video_predictor import SAM2VideoPredictor predictor = SAM2VideoPredictor.from_pretrained(avbiswas/sam2.1-hiera-tiny-mlx-4bit) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): state = predictor.init_state(<your_video>) # add new prompts and instantly get the output on the same frame frame_idx, object_ids, masks = predictor.add_new_points(state, <your_prompts>): # propagate the prompts to get masklets throughout the video for frame_idx, object_ids, masks in predictor.propagate_in_video(state): ... - Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
Upload sam2.1_hiera_tiny_image_segmenter_q8_trunk_mask_q4_memory.safetensors.json with huggingface_hub
Browse files
sam2.1_hiera_tiny_image_segmenter_q8_trunk_mask_q4_memory.safetensors.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"format": "mlx",
|
| 3 |
+
"model_id": "facebook/sam2.1-hiera-tiny",
|
| 4 |
+
"source": "checkpoints/sam2.1_hiera_tiny_image_segmenter.safetensors",
|
| 5 |
+
"variant": "mixed-q4",
|
| 6 |
+
"quantization": {
|
| 7 |
+
"bits": 4,
|
| 8 |
+
"group_size": 64,
|
| 9 |
+
"mode": "affine",
|
| 10 |
+
"recipe": "q8_trunk_mask_q4_memory"
|
| 11 |
+
},
|
| 12 |
+
"quantized_linear_count": 137
|
| 13 |
+
}
|