DeepSeek · First MLX build 17 Oct 2026
DeepSeek-V4.1-Flash-1017
MLX
TextVision mit Trending
A 289B parameter vision and text model from DeepSeek, with 8 MLX builds.
- Parameters
- 289B From Hugging Face
- Quantisation
- Unstated DeepSeek-V4.1-Flash-1017-oQ4e-mtp
- Memory needed
- 485 GB Estimate, unified memory
- Download
- 432 GB Weights, estimated
- Downloads
- 8.5K This build, last 30 days
- Task
- Vision and text Text, Vision
- MLX builds
- 8 On mlx-community
- Devices
- Mac Apple silicon
Memory
Will it fit?Needs about 485 GB at Unstated
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-vlm$ pip install -U mlx-vlm
$ python -m mlx_vlm.generate --model mlx-community/DeepSeek-V4.1-Flash-1017-oQ4e-mtp --image photo.jpg --prompt "Describe this image"MLX builds
8 on mlx-community| DeepSeek-V4.1-Flash-1017-oQ2e-mtp | Unstated | 313 GB | 353 GB | 2.2K |
|---|---|---|---|---|
| DeepSeek-V4.1-Flash-1017-oQ2.5e-mtp | Unstated | 343 GB | 386 GB | 4.1K |
| DeepSeek-V4.1-Flash-1017-oQ3e-mtp | Unstated | 381 GB | 429 GB | 1.6K |
| DeepSeek-V4.1-Flash-1017-oQ2e-ngram-mtp | Unstated | 387 GB | 435 GB | 1.0K |
| DeepSeek-V4.1-Flash-1017-oQ2.5e-ngram-mtp | Unstated | 416 GB | 468 GB | 1.4K |
| DeepSeek-V4.1-Flash-1017-oQ4e-mtp | Unstated | 432 GB | 485 GB | 8.5K |
| DeepSeek-V4.1-Flash-1017-oQ3e-ngram-mtp | Unstated | 455 GB | 511 GB | 1.9K |
| DeepSeek-V4.1-Flash-1017-oQ4e-ngram-mtp | Unstated | 506 GB | 568 GB | 3.1K |
Sizes are estimated from Hugging Face’s parameter counts and the quantisation. Downloads are for the last 30 days.