cerebras · First MLX build 5 Nov 2025
Qwen3-Coder-REAP-25B-A3B
MLX
Text apache-2.0
A 25B parameter text generation model from cerebras, with one MLX build at 8-bit.
- Parameters
- 25B From Hugging Face
- Quantisation
- 8-bit Qwen3-Coder-REAP-25B-A3B-mlx-8Bit
- Memory needed
- 32 GB Estimate, unified memory
- Download
- 26.4 GB Weights, estimated
- Downloads
- 66 This build, last 30 days
- Task
- Text generation Text
- MLX builds
- 1 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 32 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 · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/Qwen3-Coder-REAP-25B-A3B-mlx-8Bit --prompt "Hello"MLX builds
1 on mlx-community| Qwen3-Coder-REAP-25B-A3B-mlx-8Bit | 8-bit | 26.4 GB | 32 GB | 66 |
|---|
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.