Alibaba · 31B (3B active) · 專家混合（MoE）
Efficient vision MoE — 3B active, strong temporal & document understanding 看看你的 GPU 或 Mac 跑不跑得動 Qwen3-VL 30B-A3B——最低 17.3 GB，建議 28.9 GB。
2025-09256K context
專家混合（MoE）
專家總數： 128
啟用專家： 8
啟用參數： 3.0B
| 量化 | 位元 | VRAM | 品質 | 狀態 |
|---|---|---|---|---|
| Q2_K | 2 | 10.4 GB | low | — |
| Q3_K_M | 3 | 14.4 GB | moderate | — |
| Q4_K_M | 4 | 16.4 GB | good | — |
| Q5_K_M | 5 | 20.3 GB | good | — |
| Q6_K | 6 | 24.3 GB | excellent | — |
| Q8_0 | 8 | 32.3 GB | excellent | — |
| F16 | 16 | 64 GB | lossless | — |
Can I run Qwen3-VL 30B-A3B locally?
- Can I run Qwen3-VL 30B-A3B locally?
- Qwen3-VL 30B-A3B needs about 17.3 GB of memory at a minimum and 28.9 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 Qwen3-VL 30B-A3B need?
- At Q4_K_M, Qwen3-VL 30B-A3B uses about 16.4 GB of VRAM. Higher quants need more memory; lower quants fit tighter cards with a quality tradeoff.