Released: 2026-04-12
Key:
Score Threshold:
google/gemini-3-flash-preview
Score: 4346
Distance: 280km
Lat: 49.337775
Lng: 9.471616
The landscape, vegetation, and agricultural style are characteristic of Southern Germany, specifically the Baden-Württemberg region. The presence of a small service road next to rolling fields, the specific species of trees in the background, and the sky conditions are typical for this area during the summer months.
meta-llama/llama-4-maverick
meta
Score: 3540
Distance: 691km
Lat: 53.716667
Lng: -1.85
The image appears to show a rural area with a dirt path, green fields, and trees in the background. The presence of a white trailer and construction materials suggests that the area is under development or maintenance. The style of the image and the presence of a Google watermark indicate it is a Google Street View image. The landscape and vegetation are consistent with regions in Northern Europe or the UK. Based on the cloudy sky and the type of vegetation, it is likely located in the UK. The coordinates are a guess based on the typical landscape of Northern England.
anthropic/claude-haiku-4.5
anthropic
Score: 4342
Distance: 282km
Lat: 52.1
Lng: 5.8
This image shows a rural agricultural landscape with characteristic features of the Dutch countryside: flat terrain with green fields, a farmstead with typical Dutch farm buildings visible on the left, a dirt farm road, and deciduous trees lining the horizon. The landscape is extremely flat with no hills visible, typical of the Netherlands. The vegetation, agricultural style, and rural infrastructure suggest this is in the central or eastern part of the Netherlands, likely in provinces such as Gelderland, Utrecht, or Overijssel. The flat polder-like terrain and farm layout are distinctive of this region.
x-ai/grok-4-fast
x-ai
Score: 4592
Distance: 170km
Lat: 50.5
Lng: 7.5
The image depicts a rural landscape with rolling green fields, yellow crops likely rapeseed, forested hills, and an unpaved gravel road typical of agricultural areas in Western Europe. The dense woodlands and terrain suggest the Eifel region in Germany, where such scenery is common. Coordinates point to a rural spot near the Eifel National Park.
anthropic/claude-sonnet-4.6
anthropic
Score: 3777
Distance: 561km
Lat: 51.2
Lng: 13.1
The image shows a rural scene with green fields, a dirt track, construction/utility pipes (green PVC pipes), and a small industrial or farm building in the background. The landscape, vegetation, and overcast sky are typical of central/eastern Germany. The rolling green fields and forest edges, combined with the construction materials visible, suggest a rural area in Saxony or Brandenburg region of Germany.
openai/gpt-5-image-mini
openai
Score: 4737
Distance: 108km
Lat: 48.78
Lng: 6.486
Rural European landscape: narrow unpaved country lane, green cereal fields, hedgerows and mixed deciduous woodland in the background. Construction materials (green pipes) and a white trailer on the left suggest local agriculture/contractor activity rather than large-scale farming. Cloud patterns and vegetation match a temperate central-western European climate. Road and field layout, vegetation and overall feel are typical of northeastern France (Lorraine/Vosges area near Nancy), so I place the guess in that region (approx. 48.78°N, 6.486°E).
moonshotai/kimi-k2.5
moonshot
Score: 4347
Distance: 280km
Lat: 50.8234
Lng: 9.0145
The image shows a rural dirt road in a temperate European landscape with rolling hills, mixed deciduous forest, and agricultural fields. The vegetation, soil type, and farming infrastructure (including the pile of compost/manure and green drainage pipes) are characteristic of Central Europe, specifically Germany. The unpaved farm track, style of field boundaries, and forest composition match rural areas in Hesse or Lower Saxony. The presence of a white commercial vehicle and stacked materials suggests agricultural or light industrial activity common in these regions. The cloudy sky and lighting conditions are consistent with northwestern European climate patterns.
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.