meta 405b instruct cpu optimized needs about 241.7 GB of memory at its main file (a 232.7 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 6 quants, from 169.0 GB to 820.2 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 33.8 GB.
meta 405b instruct cpu optimized is a 410B-parameter Llama model in GGUF format, quantized by nisten. Its trained context length is 128K tokens. The repository has 6 quant files, from 169.0 GB to 820.2 GB (BF16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 169.0 GB (4 parts) | 177.9 GB | Too big | Get file → | |
| — | 173.0 GB (4 parts) | 181.9 GB | Too big | Get file → | |
| — | 232.7 GB (5 parts) | 241.7 GB | Too big | Get file → | |
| — | 264.0 GB (6 parts) | 273.0 GB | Too big | Get file → | |
| Q8_0 | 435.7 GB (10 parts) | 444.7 GB | Too big | Get file → | |
| BF16 | 820.2 GB (19 parts) | 829.1 GB | Too big | Get file → |
What GPU do you need for meta 405b instruct cpu optimized?
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 meta 405b instruct cpu optimized quant should you choose?
If memory is tight, the smallest file is the first one in the table at 169.0 GB, with the quality trade-offs that come with fewer bits.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about meta 405b instruct cpu optimized
How much VRAM does meta 405b instruct cpu optimized need?
About 241.7 GB at its main file with an 8K context window. That is the 232.7 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 meta 405b instruct cpu optimized on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 177.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of meta 405b instruct cpu optimized GGUF?
It depends on the quant: file is 169.0 GB, file is 173.0 GB, file is 232.7 GB, file is 264.0 GB, Q8_0 is 435.7 GB, BF16 is 820.2 GB.
What context length does meta 405b instruct cpu optimized support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 8.5 GB at 8K and 33.8 GB at 32K.
Where can I download meta 405b instruct cpu optimized GGUF?
The files are on Hugging Face in the nisten/meta-405b-instruct-cpu-optimized-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 meta 405b instruct cpu optimized 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.