Qwen3-VL-235B-A22B-Instruct
by Alibaba
· QWEN3 Vl family · released Sep 19, 2025
Open weights
Active
First open-weights release of Alibaba's Qwen3-VL vision-language line: a 235B-total/22B-active MoE with 256K context (extensible to 1M) and strong OCR/document understanding; shipped in Instruct and Thinking variants.
No pricing recorded for this model yet.
Specifications
| Model type | Multimodal |
|---|---|
| Architecture | MoE |
| Parameters | 235 (22 active) |
| Context window | 262.1K tokens |
| Max output | Not disclosed tokens |
| Knowledge cutoff | Not disclosed |
| License | Apache License 2.0 |
| Modalities | Text input Image input Video input Text output |
| Tags | Flagship Long context Vision |
| End of life | Not disclosed |
| Weights | Download |
| Last verified | Not disclosed |