DeepSeek - Mixture of experts - verified
Run DeepSeek-V4.1-Flash locally
DeepSeek-V4.1-Flash is a 763.2B mixture-of-experts (16B active) model from DeepSeek with a 1M-token context window, released under MIT (commercial use is allowed); the smallest listed download is 510.3 GB, the official Mixed FP8 (F8_E4M3), I8 and BF16 weights.
Official weights: deepseek-ai/DeepSeek-V4.1-Flash - modalities: text, image
- Total params
- 763.2B
- Active params
- 16B
- Context
- 1M
- Smallest download
- 510.3 GB
Fits in
Smallest download at most 85% of memoryNone of the tiers here. The smallest listed download is 510.3 GB, more than 85% of 256 GB.
Downloads
The official weights and every quantization listed for this model, with file sizes from Hugging Face.
| Download | Format | Size | Repo |
|---|---|---|---|
Official weightsSmallest listed | Mixed FP8 (F8_E4M3), I8 and BF16 | 510.3 GB | deepseek-ai/DeepSeek-V4.1-Flash |
4bit | MLX | 552.9 GB | mlx-community/DeepSeek-V4.1-Flash-MLX-4bit |
GB is decimal: 1 GB = 10^9 bytes. Add memory for the context and the operating system on top of the file size.
Run it
The official pages read for this directory publish no copyable run command for DeepSeek-V4.1-Flash. Download the weights from deepseek-ai/DeepSeek-V4.1-Flash and follow the model card's serving instructions.
Not sure which runtime to use? Compare local LLM runtimes by platform, API and features.
License
- License
- MIT
mit - Commercial use
- Commercial use: YesCommercial use is allowed.
- Gated download
- No. The files download directly.
Worth knowing
The card says the model activates 8B parameters per token during prefill and 16B during decode, and points to the encoding folder's Python reference implementation for building prompts.
Released (Hugging Face repo creation date).
FAQ
How big is the DeepSeek-V4.1-Flash download?
The official Mixed FP8 (F8_E4M3), I8 and BF16 weights in deepseek-ai/DeepSeek-V4.1-Flash are 510.3 GB. No smaller quantization is listed. Plan for more memory than the file size: the context (KV cache) and the operating system need room too.
Can I use DeepSeek-V4.1-Flash commercially?
DeepSeek-V4.1-Flash is released under MIT, and commercial use is allowed. The license text is at https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash.
How much memory do I need for DeepSeek-V4.1-Flash?
Its smallest listed download is 510.3 GB, more than 85% of 256 GB, so it needs a multi-GPU server or a machine with more memory than the largest tier here.
What is the context length of DeepSeek-V4.1-Flash?
1,000,000 tokens. A longer context needs more memory for the KV cache, so you can run with a shorter one when memory is tight.
Sources
Last verified from the pages below.
- Hugging Face model card - huggingface.co
- Hugging Face API - huggingface.co
- MLX 4-bit - huggingface.co
Newsletter
New open weights, every week
One email per week. New model releases, open-weight drops and what they mean for what you build.
Keep exploring
Best local AI models
Our curated picks by hardware tier, with benchmarks and the reasoning.
Model API pricing and specs
First-party prices per 1M tokens when you would rather call an API.
AI benchmark leaderboards
Sourced scores, for when you need to compare quality.
Setup guides
Point your coding tools at the model you picked.