Released: 2026-03-30
Key:
Score Threshold:
meta-llama/llama-4-maverick
meta
Score: 49
Distance: 9243km
Lat: 37.7749
Lng: -122.4194
The image is too generic and lacks distinctive features to pinpoint an exact location. However, based on the vegetation and the presence of the Google watermark, it is likely from a temperate region with moderate climate. The latitude and longitude provided are for San Francisco, California, a region known for its lush greenery and temperate climate, making it a plausible location.
x-ai/grok-4-fast
x-ai
Score: 3766
Distance: 567km
Lat: 60.1699
Lng: 24.9384
The image shows a dense stand of trees with light-colored trunks suggestive of birch trees, common in boreal forests. The lush green foliage indicates a temperate climate, likely in summer. This vegetation is typical of Northern Europe, particularly Finland, where birch forests are prevalent. Helsinki’s coordinates are used as a representative location for such scenery.
google/gemini-3-flash-preview
Score: 119
Distance: 7484km
Lat: 44.9333
Lng: -93.1867
The vegetation shown includes deciduous trees common in the Upper Midwestern United States, specifically resembling the dense foliage and silver-tinged shrubs often found along roadsides in Minnesota. The light quality and grass height are typical of late summer in a temperate climate. Given the lack of specific landmarks, I am placing this in an area with high Google Street View density and similar flora, near St. Paul, Minnesota.
anthropic/claude-haiku-4.5
anthropic
Score: 2130
Distance: 1707km
Lat: 48.8566
Lng: 2.3522
This image shows a weeping willow or similar drooping tree with characteristic hanging branches, photographed from below looking upward. The lush green foliage and the dense vegetation visible suggest a temperate climate during spring or summer. The Google Street View watermark and the natural park-like setting suggest this could be from a European location, possibly in a city park or garden area. The architectural style and vegetation patterns are consistent with Central or Western Europe. The image appears to be taken in a green space, possibly in Paris or another major European city with well-maintained parks.
anthropic/claude-sonnet-4.6
anthropic
Score: 1981
Distance: 1852km
Lat: 46.5
Lng: 2.5
The image shows dense green vegetation with what appears to be birch trees and mixed deciduous shrubs. The light quality and vegetation type suggest a temperate climate in the Northern Hemisphere during summer. The mix of tree species including what looks like birch is common in central/eastern France or similar temperate European regions. Without more distinctive features, I’ll estimate this is somewhere in central France.
openai/gpt-5-image-mini
openai
Score: 2159
Distance: 1680km
Lat: 52.2053
Lng: 0.1218
Dense roadside birch with drooping branches and an understorey of grey-green, spiny/ broom-like shrub consistent with Atlantic hedgerow species (gorse/broom). Vegetation, verge and summer foliage match rural eastern England. Google Street View imagery style also common for the UK; therefore I guess a rural location near Cambridge, England (approx. 52.2053, 0.1218).
moonshotai/kimi-k2.5
moonshot
Score: 74
Distance: 8431km
Lat: 45.5231
Lng: -122.6784
The image displays dense deciduous vegetation with lush green foliage and flowering shrubs typical of a temperate marine or oceanic climate. While the image contains no distinct geographic markers, roads, buildings, or signage, the vegetation composition—tall deciduous trees with dense understory including white-flowering shrubs—is highly characteristic of the Pacific Northwest region of North America, particularly western Oregon or Washington state. The bright lighting and full leaf coverage suggest late spring to summer in this region. However, without specific landmarks or infrastructure visible, pinpointing an exact location is impossible; this coordinate represents a reasonable estimate for the Portland, Oregon area, which shares this vegetation type and climate zone.
Each model receives the same Google Street View image. Models are tasked with guessing the location of the image using only visual clues from the street view. Their answers are evaluated without feedback or retries.