Hunyuan Models — Hardware Requirements
6 Hunyuan models from Tencent and the community, from the smallest that runs in 1.4 GB of VRAM at Q4 up to 80.4B parameters. Every row links to full quantization tables and GPU compatibility.
All Hunyuan Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| HunyuanOCR | 1.1B | 1.8 GB | 131K | ||
| Hunyuan 7B Instruct | 7.5B | 7.0 GB | 33K | ||
| Hy Memory 30B A3B | 30.1B | 19.1 GB | 262K | ||
| Hunyuan A13B Instruct | 80.4B | 50.7 GB | 33K |
Community variants and fine-tunes(2)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Hunyuan 0.5B Pretrain | 539M | 1.4 GB | 262K | ||
| Hunyuan 0.5B Instruct | 539M | 1.4 GB | 262K |
Frequently Asked Questions
- How much VRAM do I need to run a Hunyuan model?
- The smallest Hunyuan model, Hunyuan 0.5B Pretrain, runs from 1.4 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 0.6 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which Hunyuan models can I run on a 16 GB GPU?
- 4 of 6 Hunyuan models fit in 16 GB of VRAM at Q4_K_M, including HunyuanOCR, Hunyuan 7B Instruct, Hunyuan 0.5B Pretrain.
- What is the most popular Hunyuan model to run locally?
- Hy Memory 30B A3B is the most downloaded Hunyuan model in local-friendly quantized formats. It runs from 19.1 GB of VRAM at Q4_K_M.