Alibaba: qwen3-vl-plus API
An advanced multimodal large model of the Qwen3 series released by Alibaba Cloud in September 2025. Building upon the base Qwen3-VL, the Plus version is deeply optimized for Visual Agent scenarios. Beyond top-tier OCR and image comprehension, it excels at perceiving PC/Mobile GUIs to execute automation tasks like "locate and click the login button in this screenshot." Integrating DeepStack fusion and Interleaved-MRoPE, it supports hour-long video analysis with second-level temporal grounding and event summarization.
- Context window: 262,144 tokens
- Max output: 32,768 tokens
- Input: text, image
- Output: text
- Reasoning: Supported
- Tool calling: Supported
- Released: 2025-09-23
- Knowledge cutoff: 2025-03-31
Frequently Asked Questions
What is the context window of qwen3-vl-plus?
qwen3-vl-plus supports a context window of up to 262,144 tokens.
Does qwen3-vl-plus support function calling?
Yes. qwen3-vl-plus supports tool / function calling.
Does qwen3-vl-plus support reasoning?
Yes. qwen3-vl-plus is a reasoning-capable model.
What is the knowledge cutoff of qwen3-vl-plus?
The knowledge cutoff of qwen3-vl-plus is 2025-03-31.