context.vn

open Hands Index

33 models evaluated

#ModelProviderTypeScore
1Claude Fable 5AnthropicClosedclaude-fable-5
2Claude Opus 4.8AnthropicClosedclaude-opus-4-8
3Claude Opus 4.7 (Adaptive)AnthropicClosedclaude-opus-4-7
4Claude Opus 4.6Anthropic · Closed66.7%
5GPT-5.5OpenAI · Closed65.9%
6GPT-5.4OpenAI · Closed64.3%
7Gemini 3.5 FlashGoogle · Closed62.6%
8Claude Opus 4.5Anthropic · Closed60.6%
9Gemini 3.1 ProGoogle · Closed60.6%
10GPT-5.2OpenAI · Closed58.8%
11GPT-5.2-CodexOpenAI · Closed58.3%
12GLM-5.1Z.AI · Open weight58.2%
13MiniMax M3MiniMax · Open weight57.2%
14Kimi K2.6Moonshot AI · Open weight57.1%
15Claude Sonnet 4.5Anthropic · Closed53.0%
16Qwen3.6 PlusAlibaba · Closed52.9%
17GLM-5Z.AI · Open weight49.4%
18Kimi K2.5Moonshot AI · Open weight49.2%
19Gemini 3 ProGoogle · Closed49.0%
20Gemini 3 FlashGoogle · Closed49.0%
21DeepSeek V3.2 (Thinking)DeepSeek · Open weight45.7%
22MiniMax M2.5MiniMax · Closed45.2%
23Claude Sonnet 4.6Anthropic · Closed44.5%
24Qwen3 Coder NextAlibaba43.8%
25MiniMax M2.7MiniMax · Open weight43.4%
26GLM-4.7Z.AI · Open weight42.3%
27MiniMax M2.1MiniMax41.2%
28Kimi K2.5 (Reasoning)Moonshot AI · Closed41.0%
29DeepSeek V4 ProDeepSeek · Open weight40.7%
30Qwen3.5 FlashAlibaba · Closed38.1%
31Nemotron 3 Super 120B A12BNVIDIA · Open weight36.2%
32Trinity-Large-ThinkingArcee AI · Open weight32.1%
33Qwen3 Coder 480B A35BAlibaba30.9%