| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Claude Mythos 5 | Anthropic | Closed | claude-mythos-5 | |
| 2 | Claude Fable 5 | Anthropic | Closed | claude-fable-5 | |
| 3 | Claude Opus 4.8 | Anthropic | Closed | claude-opus-4-8 | |
| 4 | Gemini 3.6 Flash | Google · Closed | 83% | ||
| 5 | Holo3-35B-A3B | H Company · Open weight | 82.6% | ||
| 6 | Claude Sonnet 5 | Anthropic · Closed | 81.2% | ||
| 7 | Muse Spark 1.1 | Meta · Closed | 80.8% | ||
| 8 | Holo3-122B-A10B | H Company · Closed | 78.8% | ||
| 9 | GPT-5.5 | OpenAI · Closed | 78.7% | ||
| 10 | Gemini 3.5 Flash | Google · Closed | 78.4% | ||
| 11 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 78% | ||
| 12 | GPT-5.4 | OpenAI · Closed | 75% | ||
| 13 | Gemini 3.5 Flash-Lite | Google · Closed | 74% | ||
| 14 | Qwen3.7 Plus | Alibaba · Closed | 73.3% | ||
| 15 | Kimi K2.6 | Moonshot AI · Open weight | 73.1% | ||
| 16 | Claude Opus 4.6 | Anthropic · Closed | 72.7% | ||
| 17 | Claude Sonnet 4.6 | Anthropic · Closed | 72.1% | ||
| 18 | GPT-5.4 mini | OpenAI · Closed | 72.1% | ||
| 19 | MiniMax M3 | MiniMax · Open weight | 70.1% | ||
| 20 | Claude Opus 4.5 | Anthropic · Closed | 66.3% | ||
| 21 | GPT-5.3 Codex | OpenAI · Closed | 64.7% | ||
| 22 | Claude Sonnet 4.5 | Anthropic · Closed | 61.4% | ||
| 23 | Qwen3.5-122B-A10B | Alibaba · Open weight | 58% | ||
| 24 | Qwen3.5-27B | Alibaba · Open weight | 56.2% | ||
| 25 | Qwen3.5-35B-A3B | Alibaba · Open weight | 54.5% | ||
| 26 | GPT-5.2 | OpenAI · Closed | 47.3% | ||
| 27 | GPT-5.4 nano | OpenAI · Closed | 39% |