Released: 2026-06-26
Key:
Score Threshold:
meta-llama/llama-4-maverick
meta
Score: 4358
Distance: 275km
Lat: 52.368549
Lng: 4.882827
The image shows a parking lot with a mix of cars and a few pedestrians, set against a backdrop of trees and a blue sky. The presence of a stroller being pushed by a person and another person walking suggests a family-friendly or public area. The style of the cars and the overall setting appear to be typical of a European or Dutch environment. The presence of a ’P’ sign on a green background on the left side of the image is consistent with Dutch parking signage. The trees and greenery suggest a suburban or park-like setting. The latitude and longitude coordinates (52.368549, 4.882827) point to a location in Amsterdam, Netherlands, which matches the observed characteristics.
anthropic/claude-haiku-4.5
anthropic
Score: 4383
Distance: 264km
Lat: 52.0116
Lng: 5.0877
This appears to be a parking lot or public area in a Dutch or Northern European location based on several clues: the distinctive blue and white traffic sign on the left (characteristic of Netherlands/EU), the flat landscape with trees, the style of parked cars, the bicycle/pedestrian paths, and the overall urban planning aesthetic typical of the Netherlands. The sign format and parking area configuration suggest this is likely in a Dutch city or town, possibly near Utrecht or in the central Netherlands region.
anthropic/claude-sonnet-4.6
anthropic
Score: 3505
Distance: 711km
Lat: 58.3817
Lng: 12.3221
The image shows a parking area with Scandinavian characteristics - the style of cars, the parking signs, the landscape with deciduous trees, and the overall environment suggest Sweden. The terrain in the background with gentle hills, the style of signage (blue P parking sign), and the general atmosphere are typical of a Swedish town. The vegetation and weather suggest spring in southern/central Sweden. This looks like it could be in Västergötland region, possibly near Lidköping or a similar town in western Sweden.
openai/gpt-5-image-mini
openai
Score: 3057
Distance: 984km
Lat: 59.4015
Lng: 18.3515
Northern European look: blue ’P’ sign, compact European cars, lamp-post style and vegetation typical of Scandinavia. Scene appears to be a coastal/park parking area by a ferry/landing — family visitors with prams, map board at lot entrance. The oak-like trees, light and layout match small archipelago ferry towns around Stockholm; Vaxholm ferry/harbour parking fits these clues, so I place the guess at Vaxholm, Sweden (near the ferry/parking area).
google/gemini-3-flash-preview
Score: 4120
Distance: 387km
Lat: 48.8091803
Lng: 9.231221
The image shows a parking area near a park with European-style signage and vegetation consistent with Central Europe. Specifically, this is the parking area for the Max-Eyth-See in Stuttgart, Germany. The blue ’P’ sign and the information map on the left are typical for German public parks, and the specific layout of the parking lot and adjacent green space matches the northern section of the lake area near the Neckar river.
moonshotai/kimi-k2.5
moonshot
Score: 3813
Distance: 542km
Lat: 47.556
Lng: 10.734
Based on the visual evidence in the image, I believe this location is in Hohenschwangau, Bavaria, Germany, near Neuschwanstein Castle. Key clues include: (1) The distinctive curved black street lamp design which is characteristic of German parks and tourist areas, particularly in Bavaria; (2) The mountainous backdrop consistent with the Alpine foothills (Allgäu region); (3) The large paved parking area layout typical of major tourist destinations like Neuschwanstein Castle which attracts millions of visitors annually; (4) The deciduous trees with full foliage suggesting late spring/summer in a temperate climate; (5) The blue ’P’ parking sign and European-style vehicles parked on the left side indicating right-hand traffic; (6) The presence of pedestrians with strollers suggesting a family-friendly tourist destination. The overall scene matches the parking areas found in the Schwangau area near the famous castles.
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.