Cohere · First MLX build 25 Aug 2025
command-a-reasoning-08-2025
MLX
Text cc-by-nc-4.0
A 111B parameter text generation model from Cohere, with 2 MLX builds, from 8-bit to bf16.
- Parameters
- 111B From Hugging Face
- Quantisation
- bf16 command-a-reasoning-08-2025-bf16
- Memory needed
- 251 GB Estimate, unified memory
- Download
- 222 GB Weights, estimated
- Downloads
- 79 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 2 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 251 GB at bf16
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 · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/command-a-reasoning-08-2025-bf16 --prompt "Hello"MLX builds
2 on mlx-community| command-a-reasoning-08-2025-8bit | 8-bit | 118 GB | 134 GB | 46 |
|---|---|---|---|---|
| command-a-reasoning-08-2025-bf16 | bf16 | 222 GB | 251 GB | 79 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.