cyberagent · First MLX build 3 Jun 2026
CAT-Thinking-8B
MLX
Text apache-2.0
A 8.2B parameter text generation model from cyberagent, with 2 MLX builds, from 6-bit to 8-bit.
- Parameters
- 8.2B From Hugging Face
- Quantisation
- 8-bit CAT-Thinking-8B-MLX-8bit
- Memory needed
- 12 GB Estimate, unified memory
- Download
- 8.7 GB Weights, estimated
- Downloads
- 35 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 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 · mlx-lm$ pip install -U mlx-lm
$ mlx_lm.generate --model mlx-community/CAT-Thinking-8B-MLX-8bit --prompt "Hello"MLX builds
2 on mlx-community| CAT-Thinking-8B-MLX-6bit | 6-bit | 6.7 GB | 10 GB | 32 |
|---|---|---|---|---|
| CAT-Thinking-8B-MLX-8bit | 8-bit | 8.7 GB | 12 GB | 35 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.