| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Composer 2 | Cursor | Proprietary | 96.1% | |
| 2 | Composer 2 Fast | Cursor | Proprietary | 94.9% | |
| 3 | GPT-5.4 | OpenAI | Proprietary | 85.3% | |
| 4 | GPT-5.5 | OpenAI · Closed | 84.7% | ||
| 5 | Claude Opus 4.6 | Anthropic · Closed | 84.1% | ||
| 6 | Claude Opus 4.7 | Anthropic · Closed | 82.8% | ||
| 7 | Claude Sonnet 4.6 | Anthropic · Closed | 80.6% | ||
| 8 | Gemini 3.1 Pro | Google · Closed | 78.9% | ||
| 9 | Kimi K2.5 | Moonshot AI · Open weight | 77.2% | ||
| 10 | Gemma 4 31B | Google · Open weight | 75.2% | ||
| 11 | GLM-5 | Z.AI · Open weight | 74.8% | ||
| 12 | Grok 4 | xAI · Closed | 72.6% | ||
| 13 | GPT-OSS 120B | OpenAI · Open weight | 71.6% | ||
| 14 | DeepSeek V3.2 | DeepSeek · Open weight | 71.5% | ||
| 15 | MiniMax M2.7 | MiniMax · Open weight | 71.4% | ||
| 16 | GPT-OSS 20B | OpenAI · Open weight | 71% |