SShogo Bench

SimpleBench

Trick questions about space, time and social reasoning.

Agent Index weight
0%
shown for context only
Models tested
100
from the benchmark’s leaderboard and reports
Normalised for the index
0.1667 → 0, 1 → 100
Official site

Results

#ModelScoreNormalisedHarnessCost/taskDate
1Claude Fable 5.2 Supported
Anthropic
94.9%source——
2Gemini 4 Carbon Supported
Google
92.6%source——
3Gemini 4 Argon Supported
Google
91.7%source——
4Claude Opus 5.5 Supported
Anthropic
88.4%source——
5Claude Fable 5.1 Supported
Anthropic
86.6%source——
6GPT-6 Astra Supported
OpenAI
86.5%source——
7GPT-6.1 Sol Estimated
OpenAI
82.9%source——
8Gemini 3.8 Flash Supported
Google
82.4%source——
9Claude Fable 5 Supported
Anthropic
81.9%source——
10Muse Spark 1.3 Estimated
Meta
81.8%source——
11Claude Opus 5 Supported
Anthropic
80.6%source——
12Gemini 3.1 Pro Preview Supported
Google
79.6%source——
13GPT-5.5 Pro
OpenAI
76.9%source——
14Gemini 3.5 Flash Supported
Google
76.7%source——
15Gemini 3 Pro Preview Supported
Google
76.4%source——
16Claude Sonnet 5.5 Estimated
Anthropic
75.9%source——
17Grok 4.6 Supported
xAI
75.9%source——
18Muse Spark 1.2 Estimated
Meta
74.5%source——
19GPT-5.4 Pro
OpenAI
74.1%source——
20GPT-6 Sol Estimated
OpenAI
73.1%source——
21Muse Ember
Meta
72.7%source——
22GPT-6.1 Terra Estimated
OpenAI
72.2%source——
23GPT-5.6 Sol Supported
OpenAI
71.7%source——
24Qwen3.7 Max
Alibaba (Qwen)
70.4%source——
25Grok 4.5 Estimated
xAI
70.0%source——
26GPT-5.5 Supported
OpenAI
69.0%source——
27Claude Opus 4.6 Supported
Anthropic
67.6%source——
28DeepSeek V4.1 Flash Estimated
DeepSeek
66.7%source——
29GLM-5.3 Supported
Z.ai (Zhipu)
66.2%source——
30Claude Opus 4.8 Supported
Anthropic
64.8%source——
31Qwen3.6 Max Preview Estimated
Alibaba (Qwen)
63.0%source——
32Gemini 2.5 Pro Supported
Google
62.4%source——
33Claude Opus 4.5 (latest) Supported
Anthropic
62.0%source——
34Claude Opus 4.7 Supported
Anthropic
61.7%source——
35GPT-5 Pro
OpenAI
61.6%source——
36DeepSeek V4 Flash 0731
DeepSeek
61.1%source——
37Gemini 3 Flash Preview Supported
Google
61.1%source——
38Kimi K3 Supported
Moonshot AI
60.7%source——
39Claude Sonnet 5 Supported
Anthropic
60.6%source——
40Grok 4 Supported
xAI
60.5%source——
41Qwen3.8 27B
Alibaba (Qwen)
60.2%source——
42Claude Opus 4.1 (latest) Supported
Anthropic
60.0%source——
43Claude Haiku 5.5 Estimated
Anthropic
59.2%source——
44Claude-Opus-4 Supported
Anthropic
58.8%source——
45GLM-5.2 Supported
Z.ai (Zhipu)
58.8%source——
46Kimi K2.7 Code Estimated
Moonshot AI
57.9%source——
47GPT-5.2 Pro
OpenAI
57.4%source——
48GPT-5 Supported
OpenAI
56.7%source——
49Grok 4.1 Fast (Non-Reasoning)
xAI
56.0%source——
50GLM-5.1 Estimated
Z.ai (Zhipu)
55.1%source——
51Claude Sonnet 4.5 (latest) Supported
Anthropic
54.3%source——
52GLM-5 Supported
Z.ai (Zhipu)
53.2%source——
53GPT-5.1 Supported
OpenAI
53.2%source——
54o3 Supported
OpenAI
53.1%source——
55DeepSeek-V3.2-Speciale
DeepSeek
52.6%source——
56Inkling Estimated
thinkingmachines
50.0%source——
57GPT-5.6 Terra Estimated
OpenAI
48.9%source——
58GLM-4.7 Estimated
Z.ai (Zhipu)
47.7%source——
59GPT-5.6 Luna Estimated
OpenAI
46.8%source——
60Kimi K2.5 Supported
Moonshot AI
46.8%source——
61Claude 3.7 Sonnet Supported
Anthropic
46.4%source——
62DeepSeek V4 Flash
DeepSeek
46.3%source——
63GPT-5.2 Supported
OpenAI
45.8%source——
64MiniMax-M3 Estimated
MiniMax
45.8%source——
65Claude Sonnet 4 Supported
Anthropic
45.5%source——
66Nemotron 3 Ultra
NVIDIA
41.7%source——
67o1 Estimated
OpenAI
41.7%source——
68Claude-Sonnet-3.5 Estimated
Anthropic
41.4%source——
69Gemini 2.5 Flash Estimated
Google
41.2%source——
70DeepSeek-R1 Estimated
DeepSeek
40.8%source——
71DeepSeek V3.1
DeepSeek
40.0%source——
72Kimi K2 Thinking Estimated
Moonshot AI
39.6%source——
73o4-mini Estimated
OpenAI
38.7%source——
74Grok 3
xAI
36.1%source——
75Qwen3.6 Flash
Alibaba (Qwen)
35.2%source——
76MiniMax-M2.1
MiniMax
34.7%source——
77GPT-4.5
OpenAI
34.5%source——
78Gemini-2.0-Flash
Google
31.1%source——
79Qwen3 235B A22B
Alibaba (Qwen)
31.0%source——
80Gemini 2.0 Flash Thinking (Jan 2025)
Google
30.7%source——
81Llama 4 Maverick
Meta
27.7%source——
82DeepSeek V3 0324
DeepSeek
27.2%source——
83Gemini 1.5 Pro (Feb 2024)
Google
27.1%source——
84GPT-4.1
OpenAI
27.0%source——
85Kimi K2 0905 Estimated
Moonshot AI
26.3%source——
86GPT-4 Turbo
OpenAI
25.1%source——
87MiniMax-M2 Estimated
MiniMax
25.0%source——
88Claude 3 Opus Estimated
Anthropic
23.5%source——
89Llama 3.1-405B
Meta
23.0%source——
90o3-mini Estimated
OpenAI
22.8%source——
91Grok-2 (Dec 2024)
xAI
22.7%source——
92Mistral Large 2 (Nov 2024)
Mistral
22.5%source——
93GPT OSS 120B Supported
OpenAI
22.1%source——
94Mistral Large 3
Mistral
20.4%source——
95Llama 3.3 70B
Meta
19.9%source——
96DeepSeek-V3
DeepSeek
18.9%source——
97o1-mini
OpenAI
18.1%source——
98GPT-4o (2024-08-06) Supported
OpenAI
17.8%source——
99Command R+
Cohere
17.4%source——
100GPT-4o mini
OpenAI
10.7%source——