context.vn
Menu

sci Code

28 models evaluated

#ModelProviderTypeScore
1Sakana FuguSakana AIProprietary60.1%
2Step 5 PreviewStepFunProprietary58.9%
3Sakana Fugu-UltraSakana AIProprietary58.7%
4Qwen3.7 MaxAlibaba · Closed53.5%
5Gemini 3.5 FlashGoogle · Closed53.1%
6Kimi K2.6Moonshot AI · Open weight52.2%
7Qwen3.7 PlusAlibaba · Closed51.3%
8Inkling-SmallThinking Machines Lab · Open weight48.7%
9Kimi K2.5Moonshot AI · Open weight48.7%
10Grok 4.3xAI · Closed47.3%
11Qwen 3.6 Max (preview)Alibaba · Closed47%
12Nemotron 3 UltraNVIDIA · Open weight44.6%
13Muse Glimmer 30BMeta · Open weight43.6%
14Ling 3.0 FlashInclusionAI · Open weight41.2%
15Hy3 PreviewTencent · Open weight41.2%
16A.X K2SK Telecom · Open weight41%
17Ling 3.0 Flash FP8InclusionAI · Open weight40.4%
18Granite 4.2 30BIBM · Open weight38.8%
19Mercury 2Inception · Closed38.4%
20Mercury 2.5Inception · Closed38%
21K-EXAONE 2.0LG AI Research · Open weight37.4%
22Granite 4.2 8BIBM · Open weight36.1%
23Nemotron 3 Nano Omni 30B A3BNVIDIA · Open weight32%
24Nemotron 3.5 Lightning 30B A3B NVFP4NVIDIA · Open weight31.4%
25Agents-A1-4BInternScience · Open weight29.6%
26Ling 2.6 FlashInclusionAI · Open weight27%
27MiniCPM5-2BOpenBMB · Open weight26.3%
28Granite 4.2 3BIBM · Open weight24.1%