Released: 2026-05-05
Key:
Score Threshold:
google/gemini-3-flash-preview
Score: 3756
Distance: 572km
Lat: 51.5393043
Lng: 11.2319086
The vegetation consisting of dense, small-leaved deciduous shrubs and dry, tall grass is highly characteristic of Central European landscapes, specifically within Germany. The soil appears sandy or loamy. This specific type of roadside scrub is common in the Saxony-Anhalt region of Germany.
anthropic/claude-haiku-4.5
anthropic
Score: 3978
Distance: 457km
Lat: 45.5
Lng: 21.5
This image shows a rural, overgrown field with deciduous trees and vegetation typical of Central or Eastern Europe. The landscape appears to be abandoned agricultural land with mixed vegetation growth, characteristic of the Balkans or Eastern European regions. The vegetation density, tree types, and overall terrain suggest a location in Serbia, Romania, or a similar Central/Eastern European country. The Google Street View watermark and the general appearance of untended land with secondary growth forest indicates this is likely somewhere in the rural areas of the former Yugoslavia or Eastern European regions. Based on the vegetation patterns and landscape characteristics, I’m estimating this to be in central Serbia or a similar region in that geographic area.
anthropic/claude-sonnet-4.6
anthropic
Score: 4486
Distance: 217km
Lat: 47.2
Lng: 19.5
The image shows dense deciduous shrubs and trees with dry sandy soil in the foreground, typical of the Hungarian Great Plain (Alföld) region. The vegetation type, sandy soil, and cumulus clouds against a blue summer sky are characteristic of central Hungary. The scrubby oak/acacia bushes are very typical of the Hungarian lowlands area.
meta-llama/llama-4-maverick
meta
Score: 53
Distance: 9080km
Lat: 29.6832
Lng: -95.4033
The image appears to be a natural landscape with lush greenery, including trees and grass. The presence of a dirt patch and tall grass suggests a rural or semi-rural area. The sky is blue with white clouds, which is consistent with many regions. However, the specific combination of flora and the appearance of the vegetation suggest a location in the southern United States. The latitude and longitude provided are for a location near Houston, Texas, which matches the type of vegetation and climate suggested by the image.
x-ai/grok-4-fast
x-ai
Score: 2572
Distance: 1329km
Lat: 51.5074
Lng: -0.1278
The image shows a dense hedge row typical of rural or suburban British landscapes, with green foliage and grassy verge under a partly cloudy sky, suggesting a temperate climate like in London, UK. The style matches common Google Street View captures in the UK.
openai/gpt-5-image-mini
openai
Score: 4075
Distance: 409km
Lat: 52.15
Lng: 21
Rural lowland scene with dense mixed hedgerow (looks like blackthorn/hawthorn), sandy roadside verge and dry grasses typical of central/eastern Europe in summer. Sky and cloud type indicate a temperate continental climate. The vegetation structure and soil texture point to Poland/Polish plains rather than Mediterranean or boreal zones; I therefore place the image in the central Poland lowlands near Warsaw (around 52.15 N, 21.00 E).
moonshotai/kimi-k2.5
moonshot
Score: 77
Distance: 8359km
Lat: 39.8283
Lng: -98.5795
The image shows a generic rural scene with dense deciduous vegetation, a dirt path, and temperate climate characteristics (green summer foliage, partly cloudy sky). There are no distinctive landmarks, road signs, buildings, or unique geographic features visible that would allow precise identification. The vegetation pattern suggests a temperate zone, possibly North American Midwest or similar regions. Without distinguishing features like signage, architecture, terrain markers, or specific flora types, the exact location cannot be determined from this image alone. Coordinates provided represent a generic guess in the central United States where such rural scenery is common.
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.