DE

DeepSeek V4 Flash

DeepSeekGGUF by teamblobfishUpdated 2026-05-19
Download on Hugging Face
Parameters284B
Context1024K
Layers43
Architecturedeepseek4
Smallest quant61.5 GB
Quick answer

DeepSeek V4 Flash needs about 176.1 GB of memory at Q4_K_M (a 174.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 8 quants, from 61.5 GB (IQ1_S) to 302.3 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 2.9 GB.

DeepSeek V4 Flash – 284B parameter Mixture‑of‑Experts model with 1M‑token context

DeepSeek V4 Flash is a 284B‑parameter MoE language model with 13B activated parameters. It supports a context length of 1,048,576 tokens and is released under the MIT license. The model is optimized for long‑context efficiency with hybrid attention and manifold‑constrained hyper‑connections.

Good at

  • Long‑context text generation
  • Large‑scale knowledge retrieval
  • Multi‑domain expert routing

Limits

  • Requires V4‑aware llama.cpp fork
  • No vision or audio support
  • Large VRAM needed for full precision

Best for

  • Chatbots needing million‑token context
  • Research on long‑document summarization
  • Domain‑specific expert inference

Who it’s for. Developers building applications that need extensive context windows and multi‑expert routing. Researchers exploring MoE architectures and long‑sequence modeling will benefit from its 1M‑token capacity.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S61.5 GB (2 parts) 62.8 GB
Too big Get file →
IQ1_M64.5 GB (2 parts) 65.7 GB
Too big Get file →
IQ1_M67.9 GB (2 parts) 69.1 GB
Too big Get file →
IQ2_XXS78.5 GB (2 parts) 79.7 GB
Too big Get file →
IQ2_XS87.0 GB (2 parts) 88.2 GB
Too big Get file →
Q2_K107.0 GB (3 parts) 108.3 GB
Too big Get file →
Q4_K_M174.9 GB (4 parts) 176.1 GB
Too big Get file →
Q8_0302.3 GB (7 parts) 303.5 GB
Too big Get file →

For 8 GB VRAM, use IQ1_S (61.54 GB file, ~62.8 GB memory at 8 k context). For 16 GB VRAM, IQ2_XXS (78.52 GB file, ~79.7 GB memory) or Q2_K (107.03 GB file, ~108.3 GB memory) are suitable. Higher‑precision Q4_K_M or Q8_0 require 32 GB or more.

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?

Q4_K_M is the usual starting point for DeepSeek V4 Flash: it balances file size and quality well. Here it is a 174.9 GB file.

The Q8_0 file is 302.3 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is IQ1_S at 61.5 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturedeepseek4
Parameters284B (approximate, mixture of experts)
Layers43
Trained context1,048,576 tokens
KV cache (fp16)86 KB per token

Source

Quantized byteamblobfish
FamilyDeepSeek
Licensemit
Downloads (HF)3,789
Last updated2026-05-19

Frequently asked questions about DeepSeek V4 Flash

How much VRAM does DeepSeek V4 Flash need?

About 176.1 GB at Q4_K_M with an 8K context window. That is the 174.9 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 62.8 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: IQ1_S is 61.5 GB, IQ1_M is 64.5 GB, IQ1_M is 67.9 GB, IQ2_XXS is 78.5 GB, IQ2_XS is 87.0 GB, Q2_K is 107.0 GB and 2 more.

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 teamblobfish/DeepSeek-V4-Flash-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?

Use the IQ1_S quant, which has a file size of 61.54 GB and requires approximately 62.8 GB of VRAM at 8 k context.

Does DeepSeek V4 Flash support vision tasks?

No, the model has no vision or audio capabilities.

What backend is needed to run DeepSeek V4 Flash?

It requires the V4‑aware llama.cpp fork, which supports Apple Silicon (Metal), NVIDIA CUDA (Ada/Blackwell), and CPU.

Other GGUFs of DeepSeek-V4-Flash

Similar models

Browse all →

More from teamblobfish

All by teamblobfish →

More DeepSeek models

All DeepSeek →

Other large models

All large models →