Cogito v2 Preview Llama 109B MoE needs about 69.8 GB of memory at Q4_K_M (a 67.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 4 quants, from 53.8 GB (Q3_K_L) to 114.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 6.4 GB.
Cogito v2 Preview Llama 109B MoE is a 108B-parameter Llama model in GGUF format, quantized by lmstudio-community. Its trained context length is 256K tokens. The repository has 4 quant files, from 53.8 GB (Q3_K_L) to 114.5 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q3_K_L | 53.8 GB (2 parts) | 55.9 GB | Too big | Get file → | |
| Q4_K_M | 67.7 GB (2 parts) | 69.8 GB | Too big | Get file → | |
| Q6_K | 89.9 GB (3 parts) | 92.1 GB | Too big | Get file → | |
| Q8_0 | 114.5 GB (3 parts) | 116.6 GB | Too big | Get file → |
What GPU do you need for Cogito v2 Preview Llama 109B 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 Llama 109B MoE quant should you choose?
Q4_K_M is the usual starting point for Cogito v2 Preview Llama 109B MoE: it balances file size and quality well. Here it is a 67.7 GB file.
The Q8_0 file is 114.5 GB, about 1.7 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 53.8 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
Frequently asked questions about Cogito v2 Preview Llama 109B MoE
How much VRAM does Cogito v2 Preview Llama 109B MoE need?
About 69.8 GB at Q4_K_M with an 8K context window. That is the 67.7 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 109B MoE on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 55.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Cogito v2 Preview Llama 109B MoE GGUF?
It depends on the quant: Q3_K_L is 53.8 GB, Q4_K_M is 67.7 GB, Q6_K is 89.9 GB, Q8_0 is 114.5 GB.
What context length does Cogito v2 Preview Llama 109B MoE support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 1.6 GB at 8K and 6.4 GB at 32K.
Where can I download Cogito v2 Preview Llama 109B MoE GGUF?
The files are on Hugging Face in the lmstudio-community/cogito-v2-preview-llama-109B-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 Llama 109B MoE use?
Hugging Face lists the license as llama4. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.