| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | Claude Opus 4.8 | Anthropic | Closed | claude-opus-4-8 | |
| 2 | GPT-5.4 | OpenAI | Closed | gpt-5-4 | |
| 3 | Gemini 3.1 Pro | Closed | gemini-3-1-pro | ||
| 4 | Muse Spark | Meta · Closed | 84.1% | ||
| 5 | Claude Opus 4.6 | Anthropic · Closed | 83.1% | ||
| 6 | Qwen3.7 Plus | Alibaba · Closed | 79.0% | ||
| 7 | Gemini 3 Pro | Google · Closed | 72.7% | ||
| 8 | Holo2-235B-A22B | H Company · Open weight | 70.6% | ||
| 9 | Qwen3.6 Plus | Alibaba · Closed | 68.2% | ||
| 10 | Holo2-30B-A3B | H Company · Open weight | 66.1% | ||
| 11 | Qwen3.5 397B | Alibaba · Open weight | 65.6% | ||
| 12 | Holo2-8B | H Company · Open weight | 58.9% | ||
| 13 | Nemotron 3 Nano Omni 30B A3B | NVIDIA · Open weight | 57.8% | ||
| 14 | Holo2-4B | H Company · Open weight | 57.2% | ||
| 15 | Claude Opus 4.5 | Anthropic · Closed | 45.7% |