MultiGlobeQA (Gold Triples in Prompt): leaderboard
MultiGlobeQA (2026) is a multilingual geospatial reasoning benchmark of execution-verified questions over three knowledge graphs, covering 14 spatial-function families and 201 countries. Exact-match accuracy when the gold knowledge-graph facts are injected into the prompt, in percent; higher is better.
Metric: Exact-match accuracy (%; oracle Tier 1: the gold knowledge-graph triples behind each answer injected into the prompt as verbalized facts; exact match with answer-type tolerances (geodesic or relative tolerance, set equality, polygon IoU of at least 0.5) over the 4,979 true-premise questions of the English text small split, refusals and non-answers counted incorrect; greedy decoding, output constrained to the answer-type JSON schema; mean of three seeds). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) | 57 | #141 |
| 2 | Qwen 3.5 27B FP8 | 35.6 | #269 |
| 3 | Gemma 3 27B (IT) | 24.9 | #1306 |
| 4 | Qwen 3.5 35B A3B FP8 | 8.3 | #488 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=multiglobeqa-gold-triples-in-prompt · How It Works · Data refreshed daily, snapshot 2026-11-12.