Instructions to use midium-ai/d3-flash-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use midium-ai/d3-flash-8bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download midium-ai/d3-flash-8bit --local-dir d3-flash-8bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download quantize_run.json from midium-ai/d3-flash-8bit: direct link, hf CLI and curl.
- Browser
- Download file 262 Bytes
-
https://huggingface.co/midium-ai/d3-flash-8bit/resolve/main/quantize_run.json
- Command line
-
hf download hf://midium-ai/d3-flash-8bit/quantize_run.json
-
curl -L -o quantize_run.json https://huggingface.co/midium-ai/d3-flash-8bit/resolve/main/quantize_run.json
262 Bytes
| { | |
| "source": "vllm-sr/d3-flash@13dc1cc8ef23caefa765907e1d764751e5629f7d", | |
| "bits": 8, | |
| "group_size": 64, | |
| "method": "round-to-nearest (no calibration)", | |
| "quantized": "language model only", | |
| "tool": "scripts/decision_bakeoff/quantize_readout.py (Midium)" | |
| } |