| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | DeepSeek V4 Pro (Max) | DeepSeek | Open | deepseek-v4-pro-max | |
| 2 | DeepSeek V4 Pro (High) | DeepSeek | Open | deepseek-v4-pro-high | |
| 4 | DeepSeek V4 Flash (High) | DeepSeek · Open weight | 76.9% | ||
| 5 | Muse Spark 1.1 | Meta · Closed | 54.1% | ||
| 6 | DeepSeek V4 Pro | DeepSeek · Open weight | 44.7% | ||
| 7 | DeepSeek V4 Flash | DeepSeek · Open weight | 37.5% | ||
| 8 | Gemini 3.5 Flash | Google · Closed | 26.6% |