| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Claude Mythos 5 | Anthropic | Closed | claude-mythos-5 | |
| 2 | Claude Opus 5 | Anthropic | Closed | claude-opus-5 | |
| 3 | Muse Spark 1.1 | Meta | Closed | muse-spark-1-1 | |
| 4 | Sakana Fugu-Ultra | Sakana AI · Closed | 50% | ||
| 5 | Claude Opus 4.8 | Anthropic · Closed | 49.8% | ||
| 6 | Sakana Fugu | Sakana AI · Closed | 47.2% | ||
| 7 | Claude Opus 4.7 (Adaptive) | Anthropic · Closed | 46.9% | ||
| 8 | Gemini 3.1 Pro | Google · Closed | 45.4% | ||
| 9 | Kimi K3 | Moonshot AI · Closed | 43.5% | ||
| 10 | Claude Sonnet 5 | Anthropic · Closed | 43.2% | ||
| 11 | GPT-5.5 Pro | OpenAI · Closed | 43.1% | ||
| 12 | Muse Spark | Meta · Closed | 42.8% | ||
| 13 | GPT-5.4 Pro | OpenAI · Closed | 42.7% | ||
| 14 | GPT-5.5 | OpenAI · Closed | 41.4% | ||
| 15 | GLM-5.2 | Z.AI · Open weight | 40.5% | ||
| 16 | Claude Opus 4.6 | Anthropic · Closed | 40% | ||
| 17 | GPT-5.4 | OpenAI · Closed | 39.8% | ||
| 18 | MiMo-V2.5-Pro | Xiaomi · Closed | 34% | ||
| 19 | Grok 4.20 | xAI · Closed | 31.6% | ||
| 20 | Inkling | Thinking Machines Lab · Open weight | 30% | ||
| 21 | GPT-5.4 mini | OpenAI · Closed | 28.2% | ||
| 22 | Nemotron 3 Ultra | NVIDIA · Open weight | 26.7% | ||
| 23 | GPT-5.4 nano | OpenAI · Closed | 24.3% | ||
| 24 | Gemma 4 31B | Google · Open weight | 19.5% | ||
| 25 | Gemma 4 26B A4B | Google · Open weight | 8.7% | ||
| 26 | Gemma 4 12B | Google · Open weight | 5.2% |