SShogo Bench

GeoBench

GeoGuessr-style location from a photo.

Agent Index weight
0%
shown for context only
Models tested
42
from the benchmark’s leaderboard and reports
Normalised for the index
0 → 0, 1 → 100
Official site

Results

#ModelScoreNormalisedHarnessCost/taskDate
1Claude Fable 5 Supported
Anthropic
89.0%source——
2Gemini 3 Flash Preview Supported
Google
88.0%source——
3Gemini 2.5 Pro Supported
Google
86.0%source——
4Gemini 3.8 Flash Supported
Google
86.0%source——
5Muse Spark 1.3 Estimated
Meta
86.0%source——
6Gemini 3.5 Flash Supported
Google
85.0%source——
7Claude Fable 5.1 Supported
Anthropic
84.0%source——
8Gemini 3.1 Pro Preview Supported
Google
84.0%source——
9Gemini 3 Pro Preview Supported
Google
84.0%source——
10Gemini 3.1 Flash Lite
Google
83.0%source——
11Claude Opus 4.8 Supported
Anthropic
81.0%source——
12GPT-5 Supported
OpenAI
81.0%source——
13o1 Estimated
OpenAI
80.0%source——
14Gemini 2.0 Pro
Google
77.5%source——
15Gemini-2.0-Flash
Google
77.0%source——
16Gemini 1.5 Flash (Sep 2024)
Google
76.0%source——
17Gemini 2.5 Flash Estimated
Google
76.0%source——
18Claude Opus 4.5 (latest) Supported
Anthropic
75.0%source——
19Claude Sonnet 5 Supported
Anthropic
74.0%source——
20o3 Supported
OpenAI
74.0%source——
21Claude Sonnet 4.6 Supported
Anthropic
73.0%source——
22GPT-5.6 Sol Supported
OpenAI
73.0%source——
23GPT-4.1
OpenAI
72.0%source——
24GPT-4o (2024-08-06) Supported
OpenAI
71.0%source——
25Gemini 2.0 Flash Thinking (Jan 2025)
Google
70.0%source——
26Claude 3.7 Sonnet Supported
Anthropic
68.0%source——
27GPT-4o mini
OpenAI
64.0%source——
28o4-mini Estimated
OpenAI
64.0%source——
29Claude-Sonnet-3.5 Estimated
Anthropic
62.0%source——
30Qwen2.5-72B Estimated
Alibaba (Qwen)
62.0%source——
31Grok 4.7 Estimated
xAI
61.0%source——
32Claude Sonnet 4.5 (latest) Supported
Anthropic
60.0%source——
33Grok 4.3 Beta
xAI
56.0%source——
34Gemma 3 27B
Google
52.0%source——
35Llama 3.2 90B
Meta
52.0%source——
36Llama 4 Maverick
Meta
52.0%source——
37Claude-Opus-4 Supported
Anthropic
49.0%source——
38Grok 4 Supported
xAI
45.0%source——
39Qwen3.8 27B
Alibaba (Qwen)
38.0%source——
40Claude Sonnet 4 Supported
Anthropic
37.0%source——
41Pixtral 12B
Mistral
37.0%source——
42Claude 3.5 Haiku
Anthropic
34.0%source——