DeepSeek V4 Flash needs about 87.9 GB of memory at its main file (a 86.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 3 quants, from 86.7 GB to 86.7 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.9 GB.
DeepSeek V4 Flash – a 2150.4B parameter MoE model with one‑million token context
DeepSeek V4 Flash is a mixture‑of‑experts language model with 2150.4 billion parameters and 1048576 token context length. It is released under the MIT license and supports Hebrew language inference via Hebrus. The model is built on the DeepSeek‑4 architecture.
Good at
- Large‑context reasoning
- Multi‑expert domain specialization
- High‑parameter capacity
Limits
- No vision support
- Requires specialized Hebrus runtime
- Large memory footprint
Best for
- Long‑form content generation
- Domain‑specific expert queries
- Research on MoE architectures
Who it’s for. Researchers and developers needing a high‑parameter, long‑context language model for specialized domain tasks. The model is intended for use with the Hebrus inference engine and is not compatible with generic GGUF runners.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 86.7 GB | 87.9 GB | Too big | Get file → | |
| — | 86.7 GB | 87.9 GB | Too big | Get file → | |
| — | 86.7 GB | 87.9 GB | Too big | Get file → |
The available GGUF files are 86.72 GB each and require approximately 87.9 GB of VRAM at 8 k context. For systems with 16 GB VRAM, the model will not fit. If you have 32 GB or more, the model can be loaded with the provided files. No other quantizations are listed.
What GPU do you need for DeepSeek V4 Flash?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which DeepSeek V4 Flash quant should you choose?
If memory is tight, the smallest file is the first one in the table at 86.7 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about DeepSeek V4 Flash
How much VRAM does DeepSeek V4 Flash need?
About 87.9 GB at its main file with an 8K context window. That is the 86.7 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run DeepSeek V4 Flash on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 87.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of DeepSeek V4 Flash GGUF?
It depends on the quant: file is 86.7 GB, file is 86.7 GB, file is 86.7 GB.
What context length does DeepSeek V4 Flash support?
The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.9 GB at 32K.
Where can I download DeepSeek V4 Flash GGUF?
The files are on Hugging Face in the andreaborio/DeepSeek-V4-Flash-Hebrus-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does DeepSeek V4 Flash use?
Hugging Face lists the license as mit. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run DeepSeek V4 Flash on 8 GB VRAM?
No. The model requires about 87.9 GB of VRAM at 8 k context, far exceeding 8 GB.
Is DeepSeek V4 Flash compatible with generic llama.cpp?
No. The model uses a Hebrus‑native GGUF extension and requires the Hebrus runtime.
Does DeepSeek V4 Flash use a mixture of experts?
Yes. It is a mixture‑of‑experts model with 2150.4 B parameters.