Alibaba · 235B (22B active) · Mixture of Experts
Flagship vision-language MoE — frontier multimodal reasoning and agentic GUI control
0 downloads
0 likes
2025-11 256K context
Use Cases
chat vision reasoning code
Mixture of Experts
Total experts: 128
Active experts: 8
Active params: 22.0B
| Quant | Bits | VRAM | Quality | Status |
|---|---|---|---|---|
| Q2_K | 2 | 75.7 GB | low | — |
| Q3_K_M | 3 | 105.8 GB | moderate | — |
| Q4_K_M | 4 | 120.9 GB | good | — |
| Q5_K_M | 5 | 151 GB | good | — |
| Q6_K | 6 | 181.1 GB | excellent | — |
| Q8_0 | 8 | 241.2 GB | excellent | — |
| F16 | 16 | 482 GB | lossless | — |
GPUs that run Qwen 3 VL 235B-A22B
Qwen 3 VL 235B-A22B needs about 120.9 GB of VRAM at Q4_K_M, so it runs comfortably on a GPU with 192 GB of VRAM or more (3 of the GPUs we track). Common cards, smallest first:
About this model
Flagship vision-language MoE — frontier multimodal reasoning and agentic GUI control