Qwen 3 4B Thinking 2507
Qwen develops the Qwen 3 4B Thinking 2507, a chat model exceling at reasoning tasks such as logical reasoning, mathematics, and science, with significantly improved performance. Its enhanced capabilities are supported by a long-context understanding of up to 262,144 tokens, allowing for complex reasoning and deep understanding.
Input
Output
Context
262K
Max Output
33K
Parameters
4B
Technical Specifications
Model TypeChat
Context Window262,144 tokens
Max Output Tokens32,768 tokens
Parameters4B
Release DateAug 5, 2025
Training CutoffNot available
Licenseapache-2.0
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Resources & Links
Estimated GPU Requirements for Qwen 3 4B Thinking 2507
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
More Models by Qwen
qwen3-235b-a22b-thinking-2507262K context · 235B · Open weightsQwen 2.5 32B Instruct33K context · 32.8B · Open weightsQwen 2.5 14B Instruct AWQ33K context · 14.8B · Open weightsQwen 3 VL 8B Instruct262K context · 8.8B · Open weightsQwen 3.5 9B262K context · 9.7B · Open weightsQwen 2.5 3B Instruct33K context · 3.1B · Open weights
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Aug 8, 2026
Your AI stack shouldn't stand still.
Every month new models become cheaper, faster, and more capable. Inferbase ensures your application automatically benefits without changing a single API call.