| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Anthropic | 54.8% | Fable-5 (Claude Code) xHigh | Overall | |
| 4 | GPT-5.5 | OpenAI · Closed | 44.8% | ||
| 5 | GPT-5.4 | OpenAI · Closed | 44.3% | ||
| 6 | GLM-5.2 | Z.AI · Open weight | 42.4% | ||
| 7 | GPT-5.3 Codex | OpenAI · Closed | 42.4% | ||
| 8 | Claude Opus 4.6 | Anthropic · Closed | 35.6% | ||
| 9 | Gemini 3.1 Pro | Google · Closed | 33.8% | ||
| 10 | Claude Sonnet 4.6 | Anthropic · Closed | 32.2% | ||
| 11 | GLM-5 | Z.AI · Open weight | 24.2% | ||
| 12 | Kimi K2.5 | Moonshot AI · Open weight | 20.9% | ||
| 13 | MiniMax M2.5 | MiniMax · Closed | 19.5% | ||
| 14 | Gemini 3 Flash | Google · Closed | 10.0% |