Qwen3 VL 30B A3B Thinking
Alibaba · Multimodal · Released Oct 2025
A 30-billion-parameter multimodal vision-language model from Alibaba that processes text, images, and video with extended reasoning capabilities and a 131k-token context window.
- Strengths
- Combines visual reasoning with extended chain-of-thought processing, allowing the model to work through complex multimodal tasks step by step.
- Best for
- Multimodal tasks requiring detailed analysis of images and video alongside reasoning over the visual content, such as document interpretation, visual QA, and scene understanding with explanation.
- Limitations
- The 131k-token context window is smaller than Alibaba's dense text-only models; for tasks that are purely text-based without visual input, a non-multimodal model may be more efficient.
Input / 1M
$0.29
Output / 1M
$1.00
Cached input / 1M
—
Context window
262K
Price history
Input (solid)Output (dashed)
Price change
- 30d
- in increased 45.0% out decreased 58.3%
- 90d
- in increased 45.0% out decreased 58.3%
- 1y
- in increased 123.1% out decreased 35.9%
- Since launch
- in increased 123.1% out decreased 35.9%
Snapshots
| Effective | Input | Output | Cached in | Note | Source |
|---|---|---|---|---|---|
| 9 Oct 2026 | $0.29 | $1.00 | — | Imported from OpenRouter | openrouter.ai |
| 29 Jul 2026 | $0.2 | $2.40 | — | Imported from OpenRouter | openrouter.ai |
| 11 Jun 2026 | $0.13 | $1.56 | — | Imported from OpenRouter | openrouter.ai |