Geoguessr
Released: 2026-06-29
Key:
Score Threshold:
meta-llama/llama-4-maverick
meta
Score: 4460
Distance: 229km
Lat: 51.181612
Lng: 9.412686
The image shows a rural or semi-rural road with dense foliage and a guardrail. The road appears to be winding and is surrounded by trees and greenery. The presence of a bridge or overpass in the distance on the right side of the image suggests a hilly or mountainous terrain. The style of the guardrail and the road maintenance suggest it could be in a European country. The foliage and the climate indicated by the lush greenery and the cloud cover suggest a temperate region. Given these features, a likely location could be in central Europe, particularly in Germany, where similar landscapes are common. The coordinates (51.181612, 9.412686) point to a location near Bad Emstal in Hesse, Germany, which matches the observed scenery.
google/gemini-3-flash-preview
Score: 4232
Distance: 334km
Lat: 50.117182
Lng: 6.845612
The image features a European style roadside guardrail and road markings consistent with Western Europe. The lush, hilly green landscape with deciduous and coniferous trees, combined with the specific style of the bridge railing appearing in the bottom right, is characteristic of the Eifel region in Germany. This specific bend in the road with these vegetation types is localized to the Rhineland-Palatinate area.
anthropic/claude-haiku-4.5
anthropic
Score: 4267
Distance: 317km
Lat: 50.8
Lng: 7.4
This image shows a well-maintained curved road through a forested, hilly area with lush green vegetation characteristic of Central Europe. The road infrastructure, guardrail design, and landscape suggest this is likely in the Rhineland-Palatinate region of Germany or nearby areas. The dense deciduous and coniferous forest mix, rolling terrain, and modern road construction are typical of the Rhine valley or similar central European regions. The vegetation density and road quality point to a location in western Germany, likely in the Rhineland or similar forested areas between 50-51°N and 7-8°E.
anthropic/claude-sonnet-4.6
anthropic
Score: 4116
Distance: 389km
Lat: 50.3
Lng: 6.1
The image shows a curving road with metal guardrails, dense green vegetation, and rolling forested hills. The road style, guardrail design, and landscape are characteristic of the Ardennes region, which spans Belgium, Luxembourg, and western Germany. The lush deciduous forest and road infrastructure suggest this is likely in the Belgian or German Ardennes area, possibly near the Eifel region of Germany or the Belgian Ardennes.
openai/gpt-5-image-mini
openai
Score: 4343
Distance: 282km
Lat: 48.1765
Lng: 8.21
Temperate deciduous forest, rolling hills and narrow two‑lane road with European-style metal guardrail and white edge line suggest central Europe. Vegetation and climate look like the Black Forest region of southwestern Germany. Road curvature, low stone bridge parapet visible to the right and dense mixed woodland are typical of secondary rural roads in Baden-Württemberg, so I place the guess in the Black Forest near 48.1765 N, 8.2100 E.
moonshotai/kimi-k2.5
moonshot
Score: 4451
Distance: 233km
Lat: 49.5
Lng: 8.2
The image shows a curved rural road with a metal guardrail (Leitplanke) that is characteristic of German county roads (Landstraßen/Kreisstraßen). The dense temperate deciduous forest, road infrastructure quality, and guardrail design (horizontal beam with specific post spacing) are typical of southwestern Germany, particularly regions like the Palatinate Forest (Pfälzerwald), Black Forest (Schwarzwald), or Odenwald. The road markings and Google Street View image quality are consistent with German coverage. The hilly terrain with lush green vegetation suggests a temperate climate with adequate rainfall, which matches the Rhine-Palatinate or Baden-Württemberg regions. Without distinctive landmarks like signs or unique buildings, the exact coordinates are difficult to determine, but the general region is likely rural Germany in the 49-50°N, 7-9°E range.
Methodology Note
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.