MLX Community · First MLX build 11 Sept 2026
DeepSeek-V4.1-Flash
MLX
A 775B parameter model, with 2 MLX builds, from 2-bit to 4-bit.
- Parameters
- 775B From Hugging Face
- Quantisation
- 2-bit DeepSeek-V4.1-Flash-MLX-2bit
- Memory needed
- 269 GB Estimate, unified memory
- Download
- 239 GB Weights, estimated
- Downloads
- 907 This build, last 30 days
- Task
- —
- MLX builds
- 2 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 269 GB at 2-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/DeepSeek-V4.1-Flash-MLX-2bitMLX builds
2 on mlx-community| DeepSeek-V4.1-Flash-MLX-2bit | 2-bit | 239 GB | 269 GB | 907 |
|---|---|---|---|---|
| DeepSeek-V4.1-Flash-MLX-4bit | 4-bit | 439 GB | 493 GB | 238 |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.