g4x a22 needs about 46.3 GB of memory at its main file (a 13.6 GB file) with an 8K context window, so it fits on a 48 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 128.8 GB.
g4x a22 is a 30.7B-parameter Asura model in GGUF format, quantized by asura625. Its trained context length is 256K tokens. The repository has 1 quant file, from 13.6 GB to 13.6 GB.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 13.6 GB | 46.3 GB | Too big | Get file → |
What GPU do you need for g4x a22?
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: the best quant that fits is the main file (13.6 GB file, about 46.3 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about g4x a22
How much VRAM does g4x a22 need?
About 46.3 GB at its main file with an 8K context window. That is the 13.6 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 g4x a22 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 46.3 GB, so you would need a larger GPU or CPU offloading.
What is the file size of g4x a22 GGUF?
It depends on the quant: file is 13.6 GB.
What context length does g4x a22 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 32.2 GB at 8K and 128.8 GB at 32K.
Where can I download g4x a22 GGUF?
The files are on Hugging Face in the asura625/g4x-a22 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does g4x a22 use?
Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.