LL

Cogito-V2-Preview-Llama-405B

LlamaGGUF by unslothUpdated 2025-08-01
Download on Hugging Face
Parameters406B
Context128K
Layers126
Architecturellama
Smallest quant87.2 GB
Quick answer

Cogito-V2-Preview-Llama-405B needs about 247.8 GB of memory at Q4_K_M (a 243.1 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 26 quants, from 87.2 GB (IQ1_S) to 811.7 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 16.9 GB.

Cogito-V2-Preview-Llama-405B is a 406B-parameter Llama model in GGUF format, quantized by unsloth. Its trained context length is 128K tokens. The repository has 26 quant files, from 87.2 GB (IQ1_S) to 811.7 GB (BF16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S87.2 GB (2 parts) 91.9 GB
Too big Get file →
IQ1_M94.7 GB (2 parts) 99.5 GB
Too big Get file →
IQ2_XXS108.3 GB (3 parts) 113.0 GB
Too big Get file →
IQ2_M137.4 GB (3 parts) 142.1 GB
Too big Get file →
Q2_K149.3 GB (3 parts) 154.1 GB
Too big Get file →
Q2_K_L149.8 GB (4 parts) 154.5 GB
Too big Get file →
Q2_K_XL151.4 GB (4 parts) 156.1 GB
Too big Get file →
IQ3_XXS156.8 GB (4 parts) 161.5 GB
Too big Get file →
Q3_K_S175.2 GB (4 parts) 180.0 GB
Too big Get file →
Q3_K_M195.4 GB (4 parts) 200.1 GB
Too big Get file →
Q3_K_XL197.4 GB (4 parts) 202.1 GB
Too big Get file →
IQ4_XS216.6 GB (5 parts) 221.4 GB
Too big Get file →
IQ4_NL229.1 GB (5 parts) 233.8 GB
Too big Get file →
Q4_0229.7 GB (5 parts) 234.4 GB
Too big Get file →
Q4_K_S230.5 GB (5 parts) 235.2 GB
Too big Get file →
Q4_K_M243.1 GB (5 parts) 247.8 GB
Too big Get file →
Q4_K_XL243.7 GB (5 parts) 248.4 GB
Too big Get file →
Q4_1254.1 GB (6 parts) 258.8 GB
Too big Get file →
Q5_K_S279.3 GB (6 parts) 284.1 GB
Too big Get file →
Q5_K_XL286.5 GB (6 parts) 291.2 GB
Too big Get file →
Q5_K_M286.6 GB (6 parts) 291.4 GB
Too big Get file →
Q6_K332.9 GB (7 parts) 337.7 GB
Too big Get file →
Q6_K_XL347.4 GB (7 parts) 352.1 GB
Too big Get file →
Q8_0431.2 GB (9 parts) 436.0 GB
Too big Get file →
Q8_K_XL450.8 GB (10 parts) 455.5 GB
Too big Get file →
BF16811.7 GB (17 parts) 816.5 GB
Too big Get file →

What GPU do you need for Cogito-V2-Preview-Llama-405B?

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-Llama-405B quant should you choose?

Q4_K_M is the usual starting point for Cogito-V2-Preview-Llama-405B: it balances file size and quality well. Here it is a 243.1 GB file.

The Q8_0 file is 431.2 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_S at 87.2 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

Architecturellama
Parameters406B
Layers126
Trained context131,072 tokens
KV cache (fp16)504 KB per token

Source

Quantized byunsloth
FamilyLlama
Licensellama3.1
Downloads (HF)514
Last updated2025-08-01

Frequently asked questions about Cogito-V2-Preview-Llama-405B

How much VRAM does Cogito-V2-Preview-Llama-405B need?

About 247.8 GB at Q4_K_M with an 8K context window. That is the 243.1 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-Llama-405B on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 91.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Cogito-V2-Preview-Llama-405B GGUF?

It depends on the quant: IQ1_S is 87.2 GB, IQ1_M is 94.7 GB, IQ2_XXS is 108.3 GB, IQ2_M is 137.4 GB, Q2_K is 149.3 GB, Q2_K_L is 149.8 GB and 20 more.

What context length does Cogito-V2-Preview-Llama-405B support?

The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 4.2 GB at 8K and 16.9 GB at 32K.

Where can I download Cogito-V2-Preview-Llama-405B GGUF?

The files are on Hugging Face in the unsloth/cogito-v2-preview-llama-405B-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-Llama-405B use?

Hugging Face lists the license as llama3.1. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Other GGUFs of cogito-v2-preview-llama-405B

Similar models

Browse all →

More from unsloth

All by unsloth →

More Llama models

All Llama →

Other large models

All large models →