Released: 2026-04-21
Key:
Score Threshold:
google/gemini-3-flash-preview
Score: 4370
Distance: 269km
Lat: 48.1506461
Lng: 17.1121852
The architecture of the white buildings and the surrounding houses is characteristic of Central Europe. The vehicle types (Skoda, Volkswagen) and the street infrastructure, particularly the bicycle racks and bollards, are very common in Bratislava, Slovakia. Specifically, this location is in the Old Town area of Bratislava near the Umelka Gallery/Dostojevského rad area.
x-ai/grok-4-fast
x-ai
Score: 4378
Distance: 266km
Lat: 48.2082
Lng: 16.3738
The image shows modern white apartment buildings with flat roofs, a parking lot with European-style vehicles, and lush green trees under a partly cloudy sky, suggestive of a suburban area in Vienna, Austria. The architecture and layout match typical residential zones in Austria.
meta-llama/llama-4-maverick
meta
Score: 3916
Distance: 489km
Lat: 50.075555
Lng: 14.436388
The image appears to be a residential area with white buildings and a parking lot. The architecture and the presence of a ’Google’ watermark suggest it could be in Europe. The style of the buildings and the language not being visible but possibly being in a non-English speaking country, along with the presence of parked cars on the left side of the road, is consistent with many European countries. The specific location is likely in the Czech Republic, given the architectural style and the fact that the coordinates point to Prague.
anthropic/claude-haiku-4.5
anthropic
Score: 3171
Distance: 911km
Lat: 50.8765
Lng: 6.3882
This image shows a typical German residential street with characteristic architecture. The street features modern apartment buildings with the distinctive white/cream colored facades common in German suburbs. The street layout, parking arrangement, and tree-lined avenue are typical of suburban Germany. The architectural style of the buildings, road markings, and the overall urban planning suggest this is in North Rhine-Westphalia or a similar German region. The presence of German-style residential buildings and street furniture indicates this is likely in a mid-sized German city. Based on the architectural style and suburban layout, this appears to be in the Cologne/Aachen region of western Germany.
anthropic/claude-sonnet-4.6
anthropic
Score: 4980
Distance: 8km
Lat: 45.805
Lng: 16.01
The image shows a suburban street with white low-rise buildings, European-style cars, and lush green trees typical of Central/Eastern Europe. The architectural style, vehicle types, and general appearance suggest a Croatian or Slovenian suburban area. The building style and parking lot arrangement are consistent with Zagreb suburbs in Croatia.
openai/gpt-5-image-mini
openai
Score: 4975
Distance: 10km
Lat: 45.815
Lng: 15.9819
Central European residential scene — European-style license plates, left-side parking, common VW/Opel cars and small utility van. Architecture is low-rise white stucco and modern blocks typical of Croatian/Slovenian towns; curved green metal bollards and bike-stand style rails are commonly used in Croatian suburbs. Vegetation (broadleaf trees) and summer sky match a temperate Central Europe climate. Overall appearance best matches a residential neighborhood in Zagreb, Croatia (approximate coordinates).
moonshotai/kimi-k2.5
moonshot
Score: 4695
Distance: 126km
Lat: 46.0523
Lng: 14.5147
The image shows a residential street in what appears to be a Central European city based on the architecture, street layout, and vehicle types. The white plaster buildings with small windows, red-tiled roofs in the background, and overall urban planning are characteristic of Ljubljana, Slovenia. Specifically, this appears to be in the Moste or Bežigrad district, where similar residential buildings and curved streets with on-street parking are common. The license plates appear to be European format (white with dark text), consistent with Slovenia. The mature trees and temperate vegetation also support a location in the Ljubljana basin.
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.