4-bit Open on Hugging Face
Parameters
241M
Weights (est.)
≈ 139 MB
Mac memory (est.)
8 GB+
Size is estimated from the parameter count and quantisation. Memory allows a fifth extra for context plus 2 GB for the system; long contexts need more.
Details
- Base model
- openai/whisper-small (finetune)
- Vendor
- OpenAI · gpt-oss
- Task
- Speech recognition
- Quantisation
- 4-bit
- Licence
- apache-2.0
- Downloads
- 192 in the last 30 days · 793 all time · 0 likes
- Published
- 2025-12-14 · updated 2025-12-15 (11mo ago)
- Tags
- mlx-audio-plus, whisper, speech-recognition, speech-to-text
Other MLX builds of openai/whisper-small
| Build | Quantisation | Weights | Mac memory | Downloads |
|---|---|---|---|---|
| whisper-small-4bit (this one) | 4-bit | ≈ 139 MB | 8 GB+ | 192 |
| whisper-small-8bit | 8-bit | ≈ 258 MB | 8 GB+ | 44 |
| whisper-small-fp16 | fp16 | ≈ 481 MB | 8 GB+ | 122 |