Cogito-V2-Preview-Deepseek-671B-Moe needs about 406.5 GB of memory at Q4_K_M (a 404.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 26 quants, from 161.8 GB (TQ1_0) to 1342.3 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
Cogito‑V2‑Preview‑Deepseek‑671B‑MoE – a 671B parameter hybrid reasoning LLM
Cogito‑V2‑Preview‑Deepseek‑671B‑MoE is a 671B parameter model with mixture‑of‑experts architecture. It supports a 128k token context and is trained for reasoning, coding, STEM, and tool calling. The model is released under the MIT license for commercial use.
Good at
- Hybrid reasoning and self‑reflection
- Advanced coding and STEM tasks
- Tool‑calling capabilities
Limits
- Requires very large VRAM (hundreds of GB)
- No vision support
- Large context length may increase memory usage
Best for
- Complex reasoning workflows
- High‑level coding assistance
- Tool‑integrated applications
Who it’s for. Developers and researchers needing a large hybrid reasoning model for advanced NLP tasks. The model is suitable for applications that can allocate substantial GPU memory and require robust instruction following.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| TQ1_0 | 161.8 GB | 163.4 GB | Too big | Get file → | |
| IQ1_S | 185.9 GB (4 parts) | 187.5 GB | Too big | Get file → | |
| IQ1_M | 200.6 GB (5 parts) | 202.2 GB | Too big | Get file → | |
| IQ2_XXS | 216.8 GB (5 parts) | 218.4 GB | Too big | Get file → | |
| IQ2_M | 228.5 GB (5 parts) | 230.1 GB | Too big | Get file → | |
| Q2_K | 244.8 GB (5 parts) | 246.4 GB | Too big | Get file → | |
| Q2_K_L | 245.0 GB (5 parts) | 246.6 GB | Too big | Get file → | |
| Q2_K_XL | 251.1 GB (6 parts) | 252.7 GB | Too big | Get file → | |
| Q3_K_S | 289.7 GB (6 parts) | 291.3 GB | Too big | Get file → | |
| Q3_K_XL | 295.9 GB (7 parts) | 297.5 GB | Too big | Get file → | |
| Q3_K_M | 319.8 GB (7 parts) | 321.4 GB | Too big | Get file → | |
| IQ4_XS | 357.7 GB (8 parts) | 359.3 GB | Too big | Get file → | |
| IQ4_NL | 378.6 GB (8 parts) | 380.2 GB | Too big | Get file → | |
| Q4_0 | 379.5 GB (8 parts) | 381.1 GB | Too big | Get file → | |
| Q4_K_S | 380.5 GB (8 parts) | 382.1 GB | Too big | Get file → | |
| Q4_K_XL | 384.0 GB (8 parts) | 385.6 GB | Too big | Get file → | |
| Q4_K_M | 404.9 GB (9 parts) | 406.5 GB | Too big | Get file → | |
| Q4_1 | 420.4 GB (9 parts) | 422.0 GB | Too big | Get file → | |
| Q5_K_S | 462.2 GB (10 parts) | 463.8 GB | Too big | Get file → | |
| Q5_K_M | 475.8 GB (10 parts) | 477.4 GB | Too big | Get file → | |
| Q5_K_XL | 481.3 GB (10 parts) | 482.9 GB | Too big | Get file → | |
| Q6_K | 551.0 GB (12 parts) | 552.6 GB | Too big | Get file → | |
| Q6_K_XL | 574.9 GB (12 parts) | 576.5 GB | Too big | Get file → | |
| Q8_0 | 713.3 GB (15 parts) | 714.9 GB | Too big | Get file → | |
| Q8_K_XL | 780.6 GB (16 parts) | 782.1 GB | Too big | Get file → | |
| BF16 | 1342.3 GB (30 parts) | 1343.8 GB | Too big | Get file → |
For a 8 GB VRAM setup, the model is not feasible; the smallest quant (TQ1_0) still needs 163.4 GB memory at 8 k context. If you have a 16 GB or larger GPU, consider Q4_K_M (406.5 GB) or Q5_K_M (477.4 GB). For very large GPU clusters, Q8_0 (714.9 GB) or BF16 (1343.8 GB) are available. Memory estimates are approximate.
What GPU do you need for Cogito-V2-Preview-Deepseek-671B-Moe?
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 Cogito-V2-Preview-Deepseek-671B-Moe quant should you choose?
Q4_K_M is the usual starting point for Cogito-V2-Preview-Deepseek-671B-Moe: it balances file size and quality well. Here it is a 404.9 GB file.
The Q8_0 file is 713.3 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 TQ1_0 at 161.8 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Frequently asked questions about Cogito-V2-Preview-Deepseek-671B-Moe
How much VRAM does Cogito-V2-Preview-Deepseek-671B-Moe need?
About 406.5 GB at Q4_K_M with an 8K context window. That is the 404.9 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 Cogito-V2-Preview-Deepseek-671B-Moe on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 163.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Cogito-V2-Preview-Deepseek-671B-Moe GGUF?
It depends on the quant: TQ1_0 is 161.8 GB, IQ1_S is 185.9 GB, IQ1_M is 200.6 GB, IQ2_XXS is 216.8 GB, IQ2_M is 228.5 GB, Q2_K is 244.8 GB and 20 more.
What context length does Cogito-V2-Preview-Deepseek-671B-Moe support?
The GGUF header lists a trained context length of 163,840 tokens (160K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download Cogito-V2-Preview-Deepseek-671B-Moe GGUF?
The files are on Hugging Face in the unsloth/cogito-v2-preview-deepseek-671B-MoE-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 Cogito-V2-Preview-Deepseek-671B-Moe 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 Cogito‑V2‑Preview‑Deepseek‑671B‑MoE on 8 GB VRAM?
No. The smallest quant requires 163.4 GB memory at 8 k context, far exceeding 8 GB VRAM.
Does the model support vision tasks?
No, the model does not include vision capabilities.
What license applies to this model?
The model is released under the MIT license, allowing commercial use.