Vicuna GGUF models
158 GGUF models in the Vicuna family, quantized by 70 uploaders. Sizes run from 68M to 32.5B parameters. The longest trained context is 32K tokens.
About the Vicuna GGUF models
Vicuna is represented on ggufmodels.com by 158 GGUF models, quantized by 70 uploaders on Hugging Face. The most downloaded are vicuna 13B v1.5, vicuna 7B v1.5 and Vicuna 7b v1.3. Parameter counts run from 68M to 32.5B, and the longest trained context window is 32K tokens.
A Q4_K_M file of a Vicuna model is between 0.0 to 19.6 GB, which is the quant most people start with. Most files come from mradermacher (29), tensorblock (16) and RichardErkhov (12). The quants you will meet most often are Q2_K (85), Q4_K_M (79), Q8_0 (69) and Q3_K_M (68), where the number is how many models offer that quant.
Frequently asked questions
Which Vicuna GGUF should I download?
The most downloaded Vicuna GGUF here is vicuna 13B v1.5. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.
How much VRAM does a Vicuna model need?
Q4_K_M files of Vicuna models on this site range from 0.0 to 19.6 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.
Where do Vicuna GGUF files come from?
The most common quantizers here are mradermacher, tensorblock and RichardErkhov. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.