Geoguessr
Released: 2026-06-09
Key:
Score Threshold:
google/gemini-3-flash-preview
Score: 4453
Distance: 232km
Lat: 12.083321
Lng: -15.176412
The landscape features a combination of lush tropical trees and dry savannah grass, which is typical of the West African region. The quality of the road surface and the specific appearance of the trees, especially the large leafy Neem-like tree on the left, are frequently seen in Guinea-Bissau and neighboring Senegal. The presence of official Google Street View coverage in this rural setting is a key indicator for various countries, and the overcast lighting conditions and road quality are highly consistent with coverage found in the western part of Guinea-Bissau.
meta-llama/llama-4-maverick
meta
Score: 190
Distance: 6540km
Lat: -22.968611
Lng: 30.492778
The image appears to be a rural scene with a field and trees, and a power line running across it. The presence of tall grasses and the type of trees suggest a subtropical or tropical region. The statue in the field is likely a cultural or historical marker, which could be found in various parts of the world. However, the combination of the landscape, the type of vegetation, and the presence of a dirt road suggests a location in Africa. The coordinates -22.968611, 30.492778 correspond to an area in Mpumalanga, South Africa, which matches the observed landscape and vegetation.
anthropic/claude-haiku-4.5
anthropic
Score: 2219
Distance: 1625km
Lat: 12.5
Lng: -1.5
This image shows characteristics typical of West African savanna landscape. The vegetation pattern with scattered trees, dry grass, and red-brown soil is consistent with the Sahel region. The power lines and basic road infrastructure suggest a rural area in countries like Burkina Faso, Mali, or northern Ghana. The dry season appearance and the specific type of vegetation distribution points to the Sahel zone, approximately in central or northern West Africa, likely around Burkina Faso or Mali during the dry season.
anthropic/claude-sonnet-4.6
anthropic
Score: 2229
Distance: 1615km
Lat: 13.5
Lng: -1.5
The image shows a rural African savanna landscape with dry grass, scattered trees including what appears to be a baobab-like tree and mango trees, and a simple road. The vegetation type, dry season appearance, and overall landscape characteristics are consistent with West African savanna, likely in Burkina Faso or a neighboring country like Ghana or Mali. The combination of dry grasses, mixed tree species, and the road style suggests this is in the Sahel/Sudan savanna zone of West Africa, most likely Burkina Faso.
openai/gpt-5-image-mini
openai
Score: 2068
Distance: 1766km
Lat: 9.407
Lng: -0.853
Dry grassy roadside savanna with baobab-like dead trunk and broad-leaved trees, lateritic soil shoulder and single low-hung power line — all typical of the Sudanian zone of West Africa. Vegetation and road style suggest northern Ghana; I estimate near Tamale (approx. 9.407 N, 0.853 W).
moonshotai/kimi-k2.5
moonshot
Score: 4845
Distance: 63km
Lat: 13.4
Lng: -16
This image appears to be from West Africa, most likely The Gambia or southern Senegal. The key indicators are: 1) The tropical savanna landscape with dry grass and scattered trees including what appears to be a baobab tree; 2) The distinctive termite mound/dead tree stump in the field, which is characteristic of this region; 3) The paved rural road with dirt shoulders typical of West African infrastructure; 4) The large broad-leaved tree on the left resembles mango or other tropical fruit trees common in the Senegambia region; 5) The overhead power line and general vegetation pattern matches the coastal West African savanna belt. The coordinates provided are an approximation for the central North Bank Region of The Gambia, though it could potentially be in southern Senegal or Guinea-Bissau given the similar landscape characteristics throughout the region.
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.