Alibaba Cloud's open-source Qwen-VL edition. Usage \| API reference Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the International deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.
https://dashscope-intl.aliyuncs.com/compatible-mode/chat| Model | Context | Max out | Pricing |
|---|---|---|---|
| qwen3-0.6b | 256K | — | — |
| qwen3-1.7b | 256K | — | — |
| qwen3-14b | 256K | — | — |
| qwen3-235b-a22b-instruct | 256K | — | — |
| qwen3-235b-a22b-thinking | 256K | — | — |
| qwen3-235b-a22b | 256K | — | — |
| qwen3-30b-a3b-instruct | 256K | — | — |
| qwen3-30b-a3b-thinking | 256K | — | — |
| qwen3-30b-a3b | 256K | — | — |
| qwen3-32b | 256K | — | — |
| qwen3-4b | 256K | — | — |
| qwen3-8b | 256K | — | — |
| qwen3-asr-flash-filetrans | — | — | — |
| qwen3-asr-flash-realtime | — | — | — |
| qwen3-asr-flash | — | — | — |
| qwen3-coder-30b-a3b-instruct | 256K | — | — |
| qwen3-coder-480b-a35b-instruct | 256K | — | — |
| qwen3-coder-flash | 1M | — | — |
| qwen3-coder-next | 256K | 66K | — |
| qwen3-coder-plus | 1M | 66K | — |
| qwen3-livetranslate-flash-realtime | 53K | 4K | — |
| qwen3-livetranslate-flash | 53K | 4K | — |
| qwen3-max-preview | 256K | — | — |
| qwen3-max | 256K | 33K | — |
| qwen3-next-80b-a3b-instruct | 256K | — | — |
| qwen3-next-80b-a3b-thinking | 256K | — | — |
| qwen3-omni-flash-realtime | 66K | 16K | — |
| qwen3-omni-flash | 66K | 33K | — |
| qwen3-rerank | — | — | — |
| qwen3-tts-flash-realtime | — | — | — |
| qwen3-tts-flash | — | — | — |
| qwen3-tts-instruct-flash-realtime | — | — | — |
| qwen3-tts-instruct-flash | — | — | — |
| qwen3-vl-235b-a22b-instruct | 129K | — | — |
| qwen3-vl-30b-a3b-instruct | 129K | — | — |
| qwen3-vl-32b-instruct | 129K | — | — |
| qwen3-vl-8b-instruct | 129K | — | — |
| qwen3-vl-embedding | — | — | — |
| qwen3-vl-flash | 82K | — | — |
| qwen3-vl-plus | 262K | 82K | — |
| qwen3-vl-rerank | — | — | — |
| qwen3-vl-32b-thinking | 131K | 82K | $0.16$0.64 |
| qwen3-vl-8b-thinking | 127K | — | $0.18$2.1 |
| qwen3-vl-30b-a3b-thinking | 127K | — | $0.2$2.4 |
| qwen3-vl-235b-a22b-thinking | 127K | — | $0.4$4 |
| qwen3-omni-30b-a3b-captioner | 66K | 33K | $3.81$3.06 |
/v1/models/alibaba/qwen3-vl-235b-a22b-instructAlibaba Cloud's open-source Qwen-VL edition. Usage \| API reference Compared to Qwen2.5-VL, Qwen3-VL delivers major improvements in model capabilities: * **Agent interaction:** Operates computer or mobile interfaces. Detects GUI elements, understands functions, and calls tools to complete tasks. Achieves top-tier performance on benchmarks such as OS World. * **Visual coding:** Generates code from images or videos. Converts design mockups and website screenshots into HTML, CSS, and JavaScript code. * **Spatial intelligence:** Supports 2D and 3D localization, and accurately determines object position, viewpoint changes, and occlusion relationships. * **Long-video understanding:** Understands videos up to 20 minutes in length and pinpoints moments down to the second. * **Deep thinking:** Performs deep reasoning. Excels at spotting fine details and analyzing cause-and-effect relationships. Achieves top-tier performance on benchmarks such as MathVista and MMMU. * **OCR:** Supports 33 languages. Delivers stable performance under challenging conditions such as low light, blur, and skew. Significantly improves accuracy for rare characters, ancient script, and domain-specific terms. **Languages supported** Qwen-VL supports 33 languages: Chinese, Japanese, Korean, Indonesian, Vietnamese, Thai, English, French, German, Russian, Portuguese, Spanish, Italian, Swedish, Danish, Czech, Norwegian, Dutch, Finnish, Türkçe, Polish, Swahili, Romanian, Serbian, Greek, Kazakh, Uzbek, Cebuano, Arabic, Urdu, Persian, Hindi/Devanagari, and Hebrew. #### International If you select the International deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.