DORA - Disaster Operational Planning: leaderboard

DORA evaluates model agents answering disaster-response questions with a 108-tool geospatial library under a ReAct loop. This series reports final-answer accuracy for disaster operational planning; the five tasks cover 515 expert-authored cases across 45 disasters. Only the thirteen agent runs enter this panel; expert gold trajectories and tool-free VLM runs use different protocols. Higher is better.

Metric: Final-answer accuracy (%; ReAct agent with tools). Source: github.com. Saturation forecast: Around February 2027. 13 models tracked. Updated 2 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Flash59.82#174
2MiMo-V2-Pro55.48#329
3Grok 4.1 Fast54.23#429
4GPT-5.453.5#135
5Gemma 4 31B52.33#385
6Claude Sonnet 4.651.82#173
7Qwen 3.5 397B A17B51.82#255
8DeepSeek V3.250.57#417
9MiniMax-M2.750.16#533
10Step 3.5 Flash48.33#520
11GPT-OSS-120B42#707
12GPT-5.4 Nano39.98#645
13Qwen 3.5 35B A3B33.08#490

No result here: #1 Claude Fable 5.2, #2 GPT-6.2 Astra, #3 Gemini 4 Carbon.

Interactive version: theaggregate.ai/benchmark?slug=dora-disaster-operational-planning · How It Works · Data refreshed daily, snapshot 2026-11-17.