Vicuna Models — Hardware Requirements
3 Vicuna models from LMSYS and the community, from the smallest that runs in 6.6 GB of VRAM at Q4 up to 33B parameters. Every row links to full quantization tables and GPU compatibility.
All Vicuna Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Vicuna 7B V1.5 | 7B | 6.6 GB | 4K | ||
| Vicuna 13B V1.3 | 13B | 8.6 GB | 2K | ||
| Vicuna 33B V1.3 | 33B | 21.8 GB | 2K |
Frequently Asked Questions
- How much VRAM do I need to run a Vicuna model?
- The smallest Vicuna model, Vicuna 7B V1.5, runs from 6.6 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 3.3 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which Vicuna models can I run on a 16 GB GPU?
- 2 of 3 Vicuna models fit in 16 GB of VRAM at Q4_K_M, including Vicuna 7B V1.5, Vicuna 13B V1.3.
- What is the most popular Vicuna model to run locally?
- Vicuna 7B V1.5 is the most downloaded Vicuna model in local-friendly quantized formats. It runs from 6.6 GB of VRAM at Q4_K_M.