Models Deepseek Ai DeepSeek V2.5 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 8 quants, from 85.9 GB (Q2_K) to 167.2 GB (Q5_K_M). Longer context needs more: at 32K tokens the KV cache alone adds about 161.1 GB.
DeepSeek V2.5 – 236B parameters with tool‑calling and coding support
DeepSeek V2.5 is a 236‑billion‑parameter model that merges general chat and coding abilities. It supports function calling and JSON output. The model is available in multiple GGUF quantizations for local inference.
Good at
- Tool‑based function calling
- General chat and coding
- JSON output mode
Limits
- No vision support
- Requires large VRAM for high‑quality quantizations
- License is MIT but model license applies
Best for
- Local chat assistants
- Coding helpers with tool integration
- JSON‑centric workflows
Who it’s for. Developers and researchers who need a versatile local LLM with tool‑calling and coding capabilities. The model is suitable for building custom assistants that require structured output and external API interactions.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 85.9 GB (8 parts) | 126.7 GB | Too big | Get file → | |
| Q4_K_S | 98.1 GB (6 parts) | 138.8 GB | Too big | Get file → | |
| Q3_K_S | 101.7 GB (8 parts) | 142.4 GB | Too big | Get file → | |
| Q3_K_L | 105.9 GB (7 parts) | 146.7 GB | Too big | Get file → | |
| Q3_K_M | 112.7 GB (8 parts) | 153.4 GB | Too big | Get file → | |
| Q4_K_M | 142.5 GB (8 parts) | 183.2 GB | Too big | Get file → | |
| Q5_K_S | 162.3 GB (8 parts) | 203.1 GB | Too big | Get file → | |
| Q5_K_M | 167.2 GB (8 parts) | 208.0 GB | Too big | Get file → |
For 8 GB VRAM, use Q2_K (85.95 GB file, ~126.7 GB memory at 8 k context). For 16 GB VRAM, Q4_K_S (98.08 GB file, ~138.8 GB memory). For 32 GB VRAM, Q4_K_M (142.45 GB file, ~183.2 GB memory). Memory estimates are approximate.
What GPU do you need for Models Deepseek Ai DeepSeek V2.5?
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 Models Deepseek Ai DeepSeek V2.5 quant should you choose?
Q4_K_M is the usual starting point for Models Deepseek Ai DeepSeek V2.5: it balances file size and quality well. Here it is a 142.5 GB file.
If memory is tight, the smallest file is Q2_K at 85.9 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
Source
Frequently asked questions about Models Deepseek Ai DeepSeek V2.5
How much VRAM does Models Deepseek Ai DeepSeek V2.5 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 Models Deepseek Ai DeepSeek V2.5 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 126.7 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Models Deepseek Ai DeepSeek V2.5 GGUF?
It depends on the quant: Q2_K is 85.9 GB, Q4_K_S is 98.1 GB, Q3_K_S is 101.7 GB, Q3_K_L is 105.9 GB, Q3_K_M is 112.7 GB, Q4_K_M is 142.5 GB and 2 more.
What context length does Models Deepseek Ai DeepSeek V2.5 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 Models Deepseek Ai DeepSeek V2.5 GGUF?
The files are on Hugging Face in the MaziyarPanahi/DeepSeek-V2.5-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 Models Deepseek Ai DeepSeek V2.5 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 V2.5 on 8 GB VRAM?
Only the Q2_K quantization can be loaded, but it will still require ~126.7 GB of memory at 8 k context, so 8 GB VRAM is insufficient.
Does the model support vision?
No, the model has no vision capabilities.
What license applies to the GGUF files?
The GGUF files are released under the MIT License, but the model license applies for commercial use.