MLX Community · First MLX build 8 Mar 2024
whisper-base
MLX
Audio mit
A speech recognition model, with 6 MLX builds, from 4-bit to 2-bit.
- Parameters
- — Not published
- Quantisation
- Unstated whisper-base-mlx
- Memory needed
- — GB Estimate, unified memory
- Download
- — Weights, estimated
- Downloads
- 25K This build, last 30 days
- Task
- Speech recognition Audio
- MLX builds
- 6 On mlx-community
- Devices
- Mac Apple silicon
Memory
Hugging Face does not list a weights size for this build, so there is no memory estimate.
Estimated from the size of the weights, plus a fifth for context and working memory and 2 GB for macOS and your apps. Long contexts need more.
Run it
Terminal · Hugging Face CLI$ pip install -U huggingface_hub
$ hf download mlx-community/whisper-base-mlxMLX builds
6 on mlx-community| whisper-base-mlx | Unstated | — | — | 25K |
|---|---|---|---|---|
| whisper-base-mlx-q4 | 4-bit | — | — | 294 |
| whisper-base-mlx-4bit | 4-bit | — | — | 187 |
| whisper-base-mlx-8bit | 8-bit | — | — | 59 |
| whisper-base-mlx-fp32 | fp32 | — | — | 34 |
| whisper-base-mlx-2bit | 2-bit | — | — | 0 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.