Automatic Speech Recognition
Safetensors
MLX
mlx-audio
whisper
audio
hf-asr-leaderboard
speech-to-text
speech-to-speech
speech
speech generation
stt
Eval Results (legacy)
4-bit precision
Instructions to use mlx-community/whisper-tiny-asr-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/whisper-tiny-asr-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download mlx-community/whisper-tiny-asr-4bit --local-dir whisper-tiny-asr-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download tokenizer.json from mlx-community/whisper-tiny-asr-4bit: direct link, hf CLI and curl.
- Browser
- Download file 2.48 MB
-
https://huggingface.co/mlx-community/whisper-tiny-asr-4bit/resolve/main/tokenizer.json
- Command line
-
hf download hf://mlx-community/whisper-tiny-asr-4bit/tokenizer.json
-
curl -L -o tokenizer.json https://huggingface.co/mlx-community/whisper-tiny-asr-4bit/resolve/main/tokenizer.json
2.48 MB
File too large to display, you can check the raw version instead.