GiLLa needs about 9.4 GB of memory at its main file (a 4.7 GB file) with an 8K context window, so it fits on a 12 GB GPU.
GiLLa is a 6.74B-parameter Gilla model in GGUF format, quantized by xxidbr9. Its trained context length is 4K tokens. The repository has 1 quant file, from 4.7 GB to 4.7 GB.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 4.7 GB | 7.3 GB | Tight | Get file → |
What GPU do you need for GiLLa?
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: the best quant that fits is the main file (4.7 GB file, about 9.4 GB in use).
Specifications
Architecture from the GGUF header
Frequently asked questions about GiLLa
How much VRAM does GiLLa need?
About 9.4 GB at its main file with an 8K context window. That is the 4.7 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 GiLLa on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 9.4 GB, so you would need a larger GPU or CPU offloading.
What is the file size of GiLLa GGUF?
It depends on the quant: file is 4.7 GB.
What context length does GiLLa support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 4.3 GB at 8K and 2.1 GB at 4K.
Where can I download GiLLa GGUF?
The files are on Hugging Face in the xxidbr9/GiLLa repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.