TerraLogic - Summary Accuracy: leaderboard
Metric: Summary-generation accuracy (SummAcc) (%) in the step-by-step setting over the 545 TerraLogic optical, SAR and infrared remote-sensing tasks, each model driving the HieraPlan tool agent under a ReAct-style protocol in OpenCompass; higher is better. Source: arxiv.org. Saturation forecast: Around December 2026. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 82.83 |
| 2 | Llama 3 70B Instruct | 69.91 |
| 3 | Qwen 2.5 7B Instruct | 65.32 |
| 4 | DeepSeek V3 | 60.92 |
| 5 | GPT-3.5 | 59.08 |
| 6 | Qwen 2.5 32B Instruct | 58.9 |
| 7 | InternLM3-8B-Instruct | 50.28 |
| 8 | Gemini 2.5 Flash | 47.22 |
| 9 | Mistral 7B Instruct (v0.2) | 46.06 |
| 10 | Yi-1.5-6B-Chat | 37.98 |
| 11 | Phi-3 Mini 4K Instruct | 10.09 |
Interactive version: theaggregate.ai/benchmark?slug=terralogic-summary-accuracy · How It Works · Data refreshed daily, snapshot 2026-09-29.