Tencent · 80B (13B active) · 专家混合（MoE）
Reasoning image model — prompt rewrite, chain-of-thought and image-to-image editing 看看你的 GPU 或 Mac 跑不跑得動 HunyuanImage 3.0 Instruct——最低 44.7 GB，建議 74.5 GB。
2026-014K context
专家混合（MoE）
专家总数： 64
启用专家： 8
启用参数： 13.0B
| 量化 | 位元 | VRAM | 品質 | 状態 |
|---|---|---|---|---|
| Q2_K | 2 | 26.1 GB | low | — |
| Q3_K_M | 3 | 36.4 GB | moderate | — |
| Q4_K_M | 4 | 41.5 GB | good | — |
| Q5_K_M | 5 | 51.7 GB | good | — |
| Q6_K | 6 | 62 GB | excellent | — |
| Q8_0 | 8 | 82.5 GB | excellent | — |
| F16 | 16 | 164.4 GB | lossless | — |
Can I run HunyuanImage 3.0 Instruct locally?
- Can I run HunyuanImage 3.0 Instruct locally?
- HunyuanImage 3.0 Instruct needs about 44.7 GB of memory at a minimum and 74.5 GB recommended. Open this page to grade it against your GPU or Mac, then run it with runai, Ollama or LM Studio.
- How much VRAM does HunyuanImage 3.0 Instruct need?
- At Q4_K_M, HunyuanImage 3.0 Instruct uses about 41.5 GB of VRAM. Higher quants need more memory; lower quants fit tighter cards with a quality tradeoff.