DE

DeepSeek-Coder-V2-Lite-Instruct

DeepSeekGGUF by LoneStrikerUpdated 2024-06-17
Download on Hugging Face
Parameters15.7B
Context160K
Layers27
Architecturedeepseek2
Smallest quant8.5 GB
Quick answer

DeepSeek-Coder-V2-Lite-Instruct needs about 13.1 GB of memory at Q4_K_M (a 10.4 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 5 quants, from 8.5 GB (Q3_K_L) to 16.7 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 9.1 GB.

DeepSeek-Coder-V2-Lite-Instruct: 15.7B parameter code model

DeepSeek-Coder-V2-Lite-Instruct is a 15.7B parameter code language model based on the deepseek2 architecture. It is a Mixture-of-Experts model with a trained context of 163840 tokens. The model focuses on coding tasks and supports a wide range of programming languages.

Good at

  • Coding tasks
  • Code generation
  • Programming language support

Limits

  • No vision support
  • Limited context to 163840 tokens
  • License type is other

Best for

  • Developers needing code generation
  • Educational coding tools
  • Code review assistants

Who it’s for. This model is suitable for developers, educators, and researchers who require a powerful code generation tool. It is ideal for those who prefer an open-source alternative to large commercial code models.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q3_K_L8.5 GB 11.2 GB
Too big Get file →
Q4_K_M10.4 GB 13.1 GB
Too big Get file →
Q5_K_M11.9 GB 14.6 GB
Too big Get file →
Q6_K14.1 GB 16.8 GB
Too big Get file →
Q8_016.7 GB 19.5 GB
Too big Get file →

For an 8 GB VRAM setup, use the Q3_K_L quant (file size 8.46 GB, estimated memory 11.2 GB at 8k context). If you have 12 GB VRAM, Q4_K_M (10.36 GB, 13.1 GB) is suitable. For 16 GB VRAM, Q5_K_M (11.85 GB, 14.6 GB) or Q6_K (14.07 GB, 16.8 GB) can be used. Estimates are approximate.

What GPU do you need for DeepSeek-Coder-V2-Lite-Instruct?

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: the best quant that fits is Q3_K_L (8.5 GB file, about 11.2 GB in use).
  • 16 GB VRAM: the best quant that fits is Q5_K_M (11.9 GB file, about 14.6 GB in use).
  • 24 GB VRAM: the best quant that fits is Q8_0 (16.7 GB file, about 19.5 GB in use).

Which DeepSeek-Coder-V2-Lite-Instruct quant should you choose?

Q4_K_M is the usual starting point for DeepSeek-Coder-V2-Lite-Instruct: it balances file size and quality well. Here it is a 10.4 GB file.

The Q8_0 file is 16.7 GB, about 1.6 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 Q3_K_L at 8.5 GB, with the quality trade-offs that come with fewer bits. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturedeepseek2
Parameters15.7B (approximate, mixture of experts)
Layers27
Trained context163,840 tokens
KV cache (fp16)270 KB per token

Source

Quantized byLoneStriker
FamilyDeepSeek
Licenseother
Downloads (HF)1,759
Last updated2024-06-17

Frequently asked questions about DeepSeek-Coder-V2-Lite-Instruct

How much VRAM does DeepSeek-Coder-V2-Lite-Instruct need?

About 13.1 GB at Q4_K_M with an 8K context window. That is the 10.4 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-Lite-Instruct on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 11.2 GB, so you would need a larger GPU or CPU offloading.

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

It depends on the quant: Q3_K_L is 8.5 GB, Q4_K_M is 10.4 GB, Q5_K_M is 11.9 GB, Q6_K is 14.1 GB, Q8_0 is 16.7 GB.

What context length does DeepSeek-Coder-V2-Lite-Instruct support?

The GGUF header lists a trained context length of 163,840 tokens (160K). The KV cache grows with the context you use: about 2.3 GB at 8K and 9.1 GB at 32K.

Where can I download DeepSeek-Coder-V2-Lite-Instruct GGUF?

The files are on Hugging Face in the LoneStriker/DeepSeek-Coder-V2-Lite-Instruct-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-Lite-Instruct 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-Lite on 8 GB VRAM?

The Q3_K_L quant requires about 11.2 GB at 8k context, so 8 GB VRAM is insufficient.

What is the maximum context length for this model?

The model was trained with a context length of 163840 tokens.

Does this model support vision tasks?

No, the model does not support vision.

Similar models

Browse all →

More from LoneStriker

All by LoneStriker →

More DeepSeek models

All DeepSeek →

Other medium models

All medium models →