llava v1.6 needs about 8.9 GB of memory at its main file (a 4.1 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 2 quants, from 4.1 GB to 27.0 GB (F16).
llava v1.6 is a 6.74B-parameter LLaVA model in GGUF format, quantized by cortexso. Its trained context length is 4K tokens. The repository has 2 quant files, from 4.1 GB to 27.0 GB (F16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 4.1 GB | 6.7 GB | Fits | Get file → | |
| F16 | 27.0 GB | 29.6 GB | Too big | Get file → |
What GPU do you need for llava v1.6?
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 (4.1 GB file, about 8.9 GB in use).
- 16 GB VRAM: the best quant that fits is the main file (4.1 GB file, about 8.9 GB in use).
- 24 GB VRAM: the best quant that fits is the main file (4.1 GB file, about 8.9 GB in use).
- 32 GB VRAM: the best quant that fits is F16 (27.0 GB file, about 31.7 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about llava v1.6
How much VRAM does llava v1.6 need?
About 8.9 GB at its main file with an 8K context window. That is the 4.1 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 llava v1.6 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of llava v1.6 GGUF?
It depends on the quant: file is 4.1 GB, F16 is 27.0 GB.
What context length does llava v1.6 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 llava v1.6 GGUF?
The files are on Hugging Face in the cortexso/llava-v1.6 repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does llava v1.6 use?
Hugging Face lists the license as llama2. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.