DORA - Spatial Relational Analysis: leaderboard
Metric: Final-answer accuracy (%; ReAct agent with tools). Source: github.com. Saturation forecast: Around March 2027. 13 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 3.5 397B A17B | 53.4 |
| 2 | Grok 4.1 Fast | 53.4 |
| 3 | GPT-5.4 | 50.8 |
| 4 | DeepSeek V3.2 | 49.15 |
| 5 | Gemini 3 Flash | 49 |
| 6 | Gemma 4 31B | 49 |
| 7 | MiniMax-M2.7 | 48.53 |
| 8 | Claude Sonnet 4.6 | 48.23 |
| 9 | MiMo-V2-Pro | 47.43 |
| 10 | Step 3.5 Flash | 44.91 |
| 11 | GPT-OSS-120B | 37.58 |
| 12 | GPT-5.4 Nano | 33.41 |
| 13 | Qwen 3.5 35B A3B | 13.89 |
Interactive version: theaggregate.ai/benchmark?slug=dora-spatial-relational-analysis · How It Works · Data refreshed daily, snapshot 2026-10-02.