DeepSeek · First MLX build 5 Aug 2026
DeepSeek-V4-Flash-0731
MLX
Text mit
A 304B parameter text generation model from DeepSeek, with 2 MLX builds, from 2-bit to 4-bit.
- Parameters
- 304B From Hugging Face
- Quantisation
- 4-bit DeepSeek-V4-Flash-0731-2.4bit-mixed
- Memory needed
- 194 GB Estimate, unified memory
- Download
- 171 GB Weights, estimated
- Downloads
- 2.0K 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 194 GB at 4-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/DeepSeek-V4-Flash-0731-2.4bit-mixed --prompt "Hello"MLX builds
2 on mlx-community| DeepSeek-V4-Flash-0731-OptiQ-2bit | 2-bit | 95.3 GB | 109 GB | 835 |
|---|---|---|---|---|
| DeepSeek-V4-Flash-0731-2.4bit-mixed | 4-bit | 171 GB | 194 GB | 2.0K |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.