context.vn
Menu

mm Answer Bench

10 models evaluated

#ModelProviderTypeScore
1GLM-5.2Z.AIOpen weights91.0%
2Kimi K2.6Moonshot AIOpen weights86.0%
3Claude Opus 4.5AnthropicProprietary84.0%
4GLM-5.1Z.AI · Open weight83.8%
5Qwen3.6 PlusAlibaba · Closed83.8%
6GLM-5Z.AI · Open weight82.5%
7Kimi K2.5Moonshot AI · Open weight81.8%
8Qwen3.5 397BAlibaba · Open weight80.9%
9Qwen3.6-27BAlibaba · Open weight80.8%
10Qwen3.6-35B-A3BAlibaba · Open weight78.9%