DE

DeepSeek Coder V2 Instruct 0724

DeepSeekGGUF by bartowskiUpdated 2024-09-11
Download on Hugging Face
Parameters236B
Context160K
Layers60
Architecturedeepseek2
Smallest quant52.7 GB
Quick answer

DeepSeek Coder V2 Instruct 0724 needs about 183.2 GB of memory at Q4_K_M (a 142.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 18 quants, from 52.7 GB (IQ1_M) to 250.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 161.1 GB.

DeepSeek Coder V2 Instruct 0724 – 236B MoE code model

DeepSeek Coder V2 Instruct 0724 is a 236B parameter Mixture‑of‑Experts model focused on coding tasks. It supports 338 programming languages and a 128K context length. The model is available in multiple GGUF quantizations for local inference.

Good at

  • Code completion
  • Mathematical reasoning
  • Multi‑language support

Limits

  • Large VRAM requirement for high‑quality quants
  • No vision capabilities
  • License is other

Best for

  • Local code generation
  • Educational coding tools
  • Research on code models

Who it’s for. Developers and researchers who need a high‑performance code model that can run locally with llama.cpp. The model is suitable for those who can allocate sufficient GPU memory and are comfortable with GGUF quantizations.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_M52.7 GB (2 parts) 93.4 GB
Too big Get file →
IQ2_XXS61.5 GB (2 parts) 102.3 GB
Too big Get file →
IQ2_XS68.7 GB (2 parts) 109.5 GB
Too big Get file →
IQ2_M76.9 GB (2 parts) 117.7 GB
Too big Get file →
Q2_K85.9 GB (3 parts) 126.7 GB
Too big Get file →
Q2_K_L86.5 GB (3 parts) 127.2 GB
Too big Get file →
IQ3_XXS90.8 GB (3 parts) 131.6 GB
Too big Get file →
Q3_K_S101.7 GB (3 parts) 142.4 GB
Too big Get file →
IQ3_M103.4 GB (3 parts) 144.1 GB
Too big Get file →
Q3_K_M112.7 GB (3 parts) 153.4 GB
Too big Get file →
Q3_K_L122.4 GB (4 parts) 163.1 GB
Too big Get file →
Q3_K_XL122.8 GB (4 parts) 163.6 GB
Too big Get file →
IQ4_XS125.6 GB (4 parts) 166.3 GB
Too big Get file →
Q4_0133.4 GB (4 parts) 174.2 GB
Too big Get file →
Q4_K_M142.5 GB (4 parts) 183.2 GB
Too big Get file →
Q5_K_M167.2 GB (5 parts) 208.0 GB
Too big Get file →
Q6_K193.5 GB (5 parts) 234.3 GB
Too big Get file →
Q8_0250.6 GB (7 parts) 291.4 GB
Too big Get file →

For 8 GB VRAM, use IQ1_M (file size 52.68 GB, memory 93.4 GB at 8 k context). For 16 GB VRAM, IQ2_XS (68.71 GB, 109.5 GB). For 24 GB VRAM, Q2_K (85.95 GB, 126.7 GB). Higher quants provide better quality but need more memory.

What GPU do you need for DeepSeek Coder V2 Instruct 0724?

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 Coder V2 Instruct 0724 quant should you choose?

Q4_K_M is the usual starting point for DeepSeek Coder V2 Instruct 0724: it balances file size and quality well. Here it is a 142.5 GB file.

The Q8_0 file is 250.6 GB, about 1.8 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_M at 52.7 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

Architecturedeepseek2
Parameters236B (approximate, mixture of experts)
Layers60
Trained context163,840 tokens
KV cache (fp16)4800 KB per token

Source

Quantized bybartowski
FamilyDeepSeek
Licenseother
Downloads (HF)1,366
Last updated2024-09-11

Frequently asked questions about DeepSeek Coder V2 Instruct 0724

How much VRAM does DeepSeek Coder V2 Instruct 0724 need?

About 183.2 GB at Q4_K_M with an 8K context window. That is the 142.5 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 Coder V2 Instruct 0724 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 93.4 GB, so you would need a larger GPU or CPU offloading.

What is the file size of DeepSeek Coder V2 Instruct 0724 GGUF?

It depends on the quant: IQ1_M is 52.7 GB, IQ2_XXS is 61.5 GB, IQ2_XS is 68.7 GB, IQ2_M is 76.9 GB, Q2_K is 85.9 GB, Q2_K_L is 86.5 GB and 12 more.

What context length does DeepSeek Coder V2 Instruct 0724 support?

The GGUF header lists a trained context length of 163,840 tokens (160K). The KV cache grows with the context you use: about 40.3 GB at 8K and 161.1 GB at 32K.

Where can I download DeepSeek Coder V2 Instruct 0724 GGUF?

The files are on Hugging Face in the bartowski/DeepSeek-Coder-V2-Instruct-0724-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 Coder V2 Instruct 0724 use?

Hugging Face lists the license as other. 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 Coder V2 Instruct 0724 on 8 GB VRAM?

Use the IQ1_M quant; it requires 93.4 GB of memory at 8 k context, which is approximate and may need CPU fallback.

Does the model support image input?

No, the model has no vision capabilities.

What is the context length of this model?

The model supports up to 128K tokens of context.

Other GGUFs of DeepSeek-Coder-V2-Instruct-0724

Similar models

Browse all →

More from bartowski

All by bartowski →

More DeepSeek models

All DeepSeek →

Other large models

All large models →