XiaomiMiMo · First MLX build 11 May 2026
MiMo-V2.5-ASR
MLX
Audio apache-2.0
A 8B parameter speech recognition model from XiaomiMiMo, with 5 MLX builds, from 4-bit to fp32.
- Parameters
- 8B From Hugging Face
- Quantisation
- 8-bit MiMo-V2.5-ASR-MLX-8bit
- Memory needed
- 12 GB Estimate, unified memory
- Download
- 8.5 GB Weights, estimated
- Downloads
- 315 This build, last 30 days
- Task
- Speech recognition Audio
- MLX builds
- 5 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 12 GB at 8-bit
8 GB
16 GB
24 GB
32 GB
64 GB
128 GB
Fits with headroomTight, close other appsWon’t fit
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/MiMo-V2.5-ASR-MLX-8bitMLX builds
5 on mlx-community| MiMo-V2.5-ASR-MLX | 4-bit | 4.5 GB | 8 GB | 181 |
|---|---|---|---|---|
| MiMo-V2.5-ASR-MLX-4bit | 4-bit | 4.5 GB | 8 GB | 156 |
| MiMo-V2.5-ASR-MLX-8bit | 8-bit | 8.5 GB | 12 GB | 315 |
| MiMo-V2.5-ASR-MLX-bf16 | bf16 | 16.0 GB | 20 GB | 103 |
| MiMo-V2.5-ASR-MLX-fp32 | fp32 | 32.1 GB | 38 GB | 70 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.