Multiple Choice
MLX
Safetensors
English
Russian
qwen3_5
decision-model
vision
experimental
4-bit precision
Instructions to use ardanila/sys1mlx with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use ardanila/sys1mlx with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download ardanila/sys1mlx --local-dir sys1mlx
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download provenance.json from ardanila/sys1mlx: direct link, hf CLI and curl.
- Browser
- Download file 608 Bytes
-
https://huggingface.co/ardanila/sys1mlx/resolve/main/provenance.json
- Command line
-
hf download hf://ardanila/sys1mlx/provenance.json
-
curl -L -o provenance.json https://huggingface.co/ardanila/sys1mlx/resolve/main/provenance.json
608 Bytes
| { | |
| "base_revision": "1001bb4d826a52d1f399e183466143f4da7b741b", | |
| "adapter_revision": "70dd4088ebf4eb82d15ef57a863a5b9a98b94d6c", | |
| "text_bits": 4, | |
| "text_group_size": 64, | |
| "vision_dtype": "bfloat16", | |
| "format": "kev-vision-mlx-v1", | |
| "sha256": "68efe382da6d702a44f6f9d6ebf97788eacf023eae4ef33c732b763777443580", | |
| "bytes": 3034300803, | |
| "conversion_device": "Linux CPU MLX 0.32.2", | |
| "conversion_seconds": 439.0967882980001, | |
| "conversion_spec_sha256": "6fa932b7f75c0c820845b45a9bf4c36eedbec6648ef217552683268e422a8cc6", | |
| "base_model": "Qwen/Qwen3.5-4B-Base", | |
| "adapter_model": "jaredpalmer/kev-4b" | |
| } | |