Qwen 3 VL 4B Thinking
Qwen 3 VL 4B Thinking is a chat model developed by Qwen, capable of processing both text and image inputs, and excels in multimodal reasoning, particularly in STEM and math-related tasks, providing causal analysis and logical, evidence-based answers. Notably, it features an extended context length of up to 1M, allowing it to handle long-form content such as books and hours-long videos with full recall.
Input
Output
Context
-
Max Output
262K
Parameters
4B
Technical Specifications
Model TypeChat
Context WindowNot available
Max Output Tokens262,144 tokens
Parameters4B
Release DateOct 11, 2025
Training CutoffMar 1, 2025
Licenseapache-2.0
Open SourceYes
Input Modalities
TextImage
Output Modalities
Text
Capabilities
Benchmarks
Artificial Analysis
4.4%
HLE
49.4%
GPQA
17.1%
SciCode
70.0%
MMLU Pro
25.7
Math
6.7
Coding
32.0%
LiveCodeBench
7.9
Intelligence
Resources & Links
Estimated GPU Requirements for Qwen 3 VL 4B Thinking
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Jun 24, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.