context.vn

swe Verified

59 models evaluated

#ModelProviderTypeScore
1Claude Opus 5AnthropicClosedclaude-opus-5
2Claude Mythos 5AnthropicClosedclaude-mythos-5
3Claude Fable 5AnthropicClosedclaude-fable-5
4Claude Opus 4.8Anthropic · Closed88.6%
5Claude Opus 4.7 (Adaptive)Anthropic · Closed87.6%
6Claude Sonnet 5Anthropic · Closed85.2%
7GPT-5.3 CodexOpenAI · Closed85%
8Ornith-1.0-397BDeepReinforce AI · Open weight82.4%
9Claude Opus 4.5Anthropic · Closed80.9%
10Claude Opus 4.6Anthropic · Closed80.8%
11DeepSeek V4 Pro (Max)DeepSeek · Open weight80.6%
12MiniMax M3MiniMax · Open weight80.5%
13Qwen3.7 MaxAlibaba · Closed80.4%
14Kimi K2.6Moonshot AI · Open weight80.2%
15GPT-5.2OpenAI · Closed80%
16Claude Sonnet 4.6Anthropic · Closed79.6%
17DeepSeek V4 Pro (High)DeepSeek · Open weight79.4%
18DeepSeek V4 Flash (Max)DeepSeek · Open weight79%
19Qwen3.6 PlusAlibaba · Closed78.8%
20DeepSeek V4 Flash (High)DeepSeek · Open weight78.6%
21MiMo-V2-ProXiaomi · Closed78%
22GLM-5Z.AI · Open weight77.8%
23Qwen3.7 PlusAlibaba · Closed77.7%
24InklingThinking Machines Lab · Open weight77.6%
25Mistral Medium 3.5 128BMistral · Open weight77.6%
26Muse SparkMeta · Closed77.4%
27Qwen3.6-27BAlibaba · Open weight77.2%
28Claude Sonnet 4.5Anthropic · Closed77.2%
29Kimi K2.5Moonshot AI · Open weight76.8%
30Kimi K2.5 (Reasoning)Moonshot AI · Closed76.8%
31Grok 4.20xAI · Closed76.7%
32Qwen3.5 397BAlibaba · Open weight76.2%
33Ornith-1.0-35BDeepReinforce AI · Open weight75.6%
34MiMo-V2-OmniXiaomi · Closed74.8%
35Laguna M.1Poolside · Closed74.6%
36Claude 4.1 OpusAnthropic · Closed74.5%
37Hy3 PreviewTencent · Open weight74.4%
38GLM-4.7Z.AI · Open weight73.8%
39DeepSeek V4 FlashDeepSeek · Open weight73.7%
40DeepSeek V4 ProDeepSeek · Open weight73.6%
41MAI-Thinking-1Microsoft · Closed73.5%
42Qwen3.6-35B-A3BAlibaba · Open weight73.4%
43MiMo-V2-FlashXiaomi · Open weight73.4%
44Claude Haiku 4.5Anthropic · Closed73.3%
45Claude 4 SonnetAnthropic · Closed72.7%
46Qwen3.5-27BAlibaba · Open weight72.4%
47Qwen3.5-122B-A10BAlibaba · Open weight72%
48Nemotron 3 UltraNVIDIA · Open weight71.9%
49Grok Code Fast 1xAI · Closed70.8%
50Laguna XS.2Poolside · Open weight69.9%
51Ornith-1.0-9BDeepReinforce AI · Open weight69.4%
52Qwen3.5-35B-A3BAlibaba · Open weight69.2%
53Gemini 2.5 ProGoogle · Closed63.8%
54GPT-4.1OpenAI · Closed54.6%
55ZAYA1-74B-PreviewZyphra · Open weight53.2%
56o3-miniOpenAI · Closed49.3%
57Claude 3.5 SonnetAnthropic · Closed49%
58DeepSeek V3DeepSeek · Open weight42%
59GPT-4.1 miniOpenAI · Closed23.6%