GeoBrowse - Level 1 State (Direct): leaderboard

GeoBrowse (2026) Level 1: models compose weak visual cues in a street scene into a location. This board scores the state-level targets with the model answering directly without tools; accuracy in percent from an LLM judge, mean of three runs, higher is better.

Metric: Accuracy (%) on the state-level targets of GeoBrowse Level 1 (199 street-scene instances whose location must be composed from weak, fragmented visual cues), judged correct by an LLM judge with the Humanity's Last Exam judging prompt, mean of 3 runs, under direct inference: the model answers in a single pass without tool calls; higher is better. Source: arxiv.org. Saturation forecast: Around April 2028. 12 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro27.3#146
2Claude Opus 4.523.6#165
3Gemini 2.5 Pro21.8#280
4GPT-521.8#185
5Gemini 2.5 Flash18.2#466
6GPT-4o16.4#754
7Qwen 3 VL 32B16.4#624
8GPT-4.114.5#506
9Llama 3.2 90B10.9#1761
10GPT-5.29.1#190
11Qwen 2.5 VL 32B9.1#944
12Qwen 2.5 VL 7B6.3#1878

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

Interactive version: theaggregate.ai/benchmark?slug=geobrowse-level-1-state-direct · How It Works · Data refreshed daily, snapshot 2026-11-22.