测试版
排行榜/Qwen3 4B Thinking 2507
Alibaba logo

Qwen3 4B Thinking 2507

开源

来自 Alibaba · 发布于 2025-08-05

60.6
平均分
N/A
输入价格
N/A
输出价格
N/A
上下文窗口
text-generation
类型

Tested on 6 benchmarks with 60.6% average. Top scores: OpenCompass — IFEval (88.5%), OpenCompass — AIME2025 (80.0%), OpenCompass — MMLU-Pro (72.8%).

基准测试类别分数Bar
OpenCompass — IFEvallanguage88.5
OpenCompass — AIME2025math80.0
OpenCompass — MMLU-Proknowledge72.8
OpenCompass — GPQA-Diamondknowledge64.7
OpenCompass — LiveCodeBenchV6coding51.6
OpenCompass — HLEknowledge6.0