context.vn
Menu

super Gpqa

20 models evaluated

#ModelProviderTypeScore
1Claude Opus 4.6AnthropicProprietary95%
2Claude Sonnet 4.6AnthropicProprietary95%
3Qwen 3.6 Max (preview)AlibabaProprietary73.9%
4Qwen3.7 MaxAlibaba · Closed73.6%
5Qwen3.6 PlusAlibaba · Closed71.6%
6Qwen3.7 PlusAlibaba · Closed71.4%
7Seed 2.1 ProByteDance · Closed70.8%
8Claude Opus 4.5Anthropic · Closed70.6%
9Qwen3.5 397BAlibaba · Open weight70.4%
10Kimi K2.5Moonshot AI · Open weight69.2%
11Seed 2.1 TurboByteDance · Closed67.4%
12Qwen3.5-122B-A10BAlibaba · Open weight67.1%
13GLM-5Z.AI · Open weight66.8%
14Qwen3.6-27BAlibaba · Open weight66%
15Qwen3.5-27BAlibaba · Open weight65.6%
16Qwen3.6-35B-A3BAlibaba · Open weight64.7%
17Qwen3.5-35B-A3BAlibaba · Open weight63.4%
18Qwen3 235B 2507Alibaba · Open weight62.6%
21MiniCPM5-2BOpenBMB · Open weight40.8%
22MiniCPM5-1BOpenBMB · Open weight23.1%