context.vn
Menu

office Qa Pro

12 models evaluated

#ModelProviderTypeScore
1Claude Opus 5.5AnthropicProprietary67.7%
2Claude Opus 5AnthropicProprietary66.9%
3Claude Opus 4.8AnthropicProprietary66.2%
4Hy4 previewTencent · Open weight66.2%
5Kimi K3Moonshot AI · Closed63.3%
6GLM-5.3-FlashZ.AI · Open weight62.4%
7Step 5 PreviewStepFun · Closed60.3%
8Claude Fable 5Anthropic · Closed57.9%
9GPT-5.5OpenAI · Closed54.1%
10GPT-5.4OpenAI · Closed53.2%
11MiniMax M3MiniMax · Open weight45.1%
12Claude Opus 4.7 (Adaptive)Anthropic · Closed43.6%