DeepSeek-V4-Pro-0813 needs about 851.2 GB of memory at Q4_K_XL (a 849.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 2 quants, from 849.7 GB (Q4_K_XL) to 873.4 GB (Q8_K_XL). Longer context needs more: at 32K tokens the KV cache alone adds about 4.1 GB.
DeepSeek-V4-Pro-0813 – 1.57T parameter model with DSpark speculative decoding
DeepSeek-V4-Pro-0813 is a 1.57T parameter model built on the deepseek4 architecture. It includes a DSpark speculative decoding module and is released under the MIT license. The model supports a 1,048,576 token context window.
Good at
- Large context handling
- Agentic reasoning
- High parameter count
Limits
- Requires large VRAM – 851.2 GB at 8 k context for Q4_K_XL
- No vision support
- No explicit language list
Best for
- Long‑form content generation
- Complex coding tasks
- Large‑scale inference
Who it’s for. Developers and researchers needing a high‑capacity model for advanced reasoning and coding tasks. The model is suitable for environments with ample GPU memory and the ability to run GGUF weights via llama.cpp or vLLM.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_XL | 849.7 GB (20 parts) | 851.2 GB | Too big | Get file → | |
| Q8_K_XL | 873.4 GB (20 parts) | 875.0 GB | Too big | Get file → |
For 8 GB VRAM, the Q4_K_XL quant is the only available option, but it will exceed memory limits. The Q8_K_XL quant is larger and also exceeds typical consumer GPU memory. Users with multi‑GPU setups should consider the Q4_K_XL quant for best trade‑off between size and performance.
What GPU do you need for DeepSeek-V4-Pro-0813?
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-Pro-0813 quant should you choose?
Q4_K_XL (849.7 GB) is the middle option in this repository. The usual sweet spot between file size and quality, and the quant most people start with.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about DeepSeek-V4-Pro-0813
How much VRAM does DeepSeek-V4-Pro-0813 need?
About 851.2 GB at Q4_K_XL with an 8K context window. That is the 849.7 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-Pro-0813 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 851.2 GB, so you would need a larger GPU or CPU offloading.
What is the file size of DeepSeek-V4-Pro-0813 GGUF?
It depends on the quant: Q4_K_XL is 849.7 GB, Q8_K_XL is 873.4 GB.
What context length does DeepSeek-V4-Pro-0813 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 1.0 GB at 8K and 4.1 GB at 32K.
Where can I download DeepSeek-V4-Pro-0813 GGUF?
The files are on Hugging Face in the unsloth/DeepSeek-V4-Pro-0813-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-Pro-0813 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-Pro-0813 on 8 GB VRAM?
No. The smallest quant requires 851.2 GB at 8 k context, far beyond 8 GB VRAM.
Does DeepSeek-V4-Pro-0813 support vision tasks?
No, the model does not include vision capabilities.