context.vn
Menu

live Code Bench V6

31 models evaluated

#ModelProviderTypeScore
1Sakana Fugu-UltraSakana AIProprietary93.2%
2Sakana FuguSakana AIProprietary92.9%
3Qwen3.8-Omni-FlashAlibabaProprietary92.6%
4Solar Open 2Upstage · Open weight92.4%
5Qwen3.8-Flash-NextAlibaba · Open weight91.9%
6dots3-note PreviewDots Studio · Open weight91.5%
7Qwen3.8-27BAlibaba · Open weight90.3%
8Ternary Bonsai 2 27BPrism ML · Open weight90.1%
9Kimi K2.6Moonshot AI · Open weight89.6%
10Nemotron 3 UltraNVIDIA · Open weight89.0%
11BTL-3Bad Theory Labs · Open weight88.1%
12MAI-Thinking-1Microsoft · Closed87.7%
13Qwen3.6 PlusAlibaba · Closed87.1%
14Kimi K2.5Moonshot AI · Open weight85.0%
15Claude Opus 4.5Anthropic · Closed84.8%
16A.X K2SK Telecom · Open weight84.0%
17Qwen3.5 397BAlibaba · Open weight83.6%
18Granite 4.2 30BIBM · Open weight75.8%
19Granite 4.2 8BIBM · Open weight73.2%
20Gemma 4 12BGoogle · Open weight72.0%
21Mellum2-12B-A2.5B-ThinkingJetBrains · Open weight69.9%
22Granite 4.2 3BIBM · Open weight69.7%
23MiniCPM5-2BOpenBMB · Open weight69.1%
24BTL-4Bad Theory Labs · Open weight66.1%
25ZAYA1-8BZyphra · Open weight65.8%
26ZAYA1-74B-PreviewZyphra · Open weight65.7%
27Agents-A1-4BInternScience · Open weight59.6%
28LFM2.5-2.6BLiquidAI · Open weight59.4%
29Mellum2-12B-A2.5B-InstructJetBrains · Open weight37.2%
30MiniCPM5-1BOpenBMB · Open weight33.5%
31LLaDA2.2-miniInclusionAI · Open weight28.1%