Mystery Game Puzzles

Novel logic puzzles built from unfamiliar game rules. Run by Epoch AI.

CategoryReasoning
MetricAccuracy
SourceEpoch AI (evaluator)
LicenseCC BY 4.0
Benchmark pageepoch.ai
StatusScored
Weight0.8 within Reasoning
FitDiscrimination 0.91, noise 0.64

Results

#ModelProviderResultDate
1 Claude Fable 5.2 Max Anthropic 90.0% 2026-11-06
2 GPT-6 Astra Max OpenAI 84.0% 2026-08-30
3 GPT-6.1 Sol Max OpenAI 80.0% 2026-09-29
4 GPT-6.1 Terra Max OpenAI 74.0% 2026-10-26
5 Gemini 4 Carbon High Google 73.0% 2026-11-02
6 Claude Opus 5.5 Max Anthropic 71.0% 2026-09-29
7 Claude Sonnet 5.5 Max Anthropic 65.0% 2026-09-29
8 Gemini 4 Argon High Google 64.0% 2026-10-13
9 Claude Opus 5 Max Anthropic 59.0% 2026-07-25
10 GPT-5.6 Sol Max OpenAI 58.0% 2026-07-28
11 Claude Fable 5.1 Max Anthropic 58.0% 2026-09-01
12 GPT-6 Sol Max OpenAI 56.0% 2026-09-22
13 GPT-5.5 Extra High OpenAI 56.0% 2026-07-24
14 Claude Fable 5 Max Anthropic 52.0% 2026-07-31
15 GPT-5.5 High OpenAI 52.0% 2026-07-27
16 Muse Wonder Max Meta 51.0% 2026-10-14
17 DeepSeek V4.1 Flash 1017 Max DeepSeek 49.0% 2026-10-19
18 Gemini 3.8 Flash High Google 47.0% 2026-09-02
19 DeepSeek V4.1 Flash Max DeepSeek 43.0% 2026-10-08
20 DeepSeek V4 Pro 0813 Max DeepSeek 43.0% 2026-08-19
21 Kimi K3.1 Pro Max Moonshot AI 42.0% 2026-11-06
22 Qwen3.8 Max Extra High Alibaba 38.0% 2026-08-05
23 Claude Opus 5 Anthropic 37.0% 2026-08-06
24 Gemini 3.7 Flash High Google 37.0% 2026-08-14
25 GPT-5.4 Extra High OpenAI 37.0% 2026-07-24
26 Claude Opus 4.8 Max Anthropic 36.0% 2026-07-25
27 Claude Sonnet 5 Max Anthropic 35.0% 2026-07-28
28 GPT-5.6 Terra Max OpenAI 35.0% 2026-07-28
29 Grok 4.6 Extra High xAI 34.0% 2026-08-14
30 Gemini 3.1 Pro Preview High Google 34.0% 2026-07-27
31 DeepSeek V4 Flash 0731 Max DeepSeek 34.0% 2026-08-05
32 GLM-5.3 Max Zhipu AI 33.0% 2026-08-30
33 GPT-5.6 Sol Non-reasoning OpenAI 33.0% 2026-08-27
34 Qwen3.7 Max Alibaba 32.0% 2026-07-28
35 Gemini 3.5 Flash High Google 32.0% 2026-07-27
36 Gemini 3.1 Pro Preview Medium Google 32.0% 2026-08-06
37 Claude Opus 4.8 Extra High Anthropic 31.0% 2026-07-26
38 Claude Haiku 5.5 Max Anthropic 30.0% 2026-10-07
39 Gemini 3.6 Flash High Google 30.0% 2026-08-05
40 Grok 4.7 Extra High xAI 29.0% 2026-09-22
41 o3 High OpenAI 29.0% 2026-08-27
42 Gemini 3.1 Pro Preview Low Google 29.0% 2026-08-06
43 Gemini 3.5 Flash Low Google 28.0% 2026-08-05
44 Claude Opus 4.7 Max Anthropic 28.0% 2026-07-25
45 GPT-5.4 Medium OpenAI 28.0% 2026-08-28
46 GPT-5.5 Low OpenAI 28.0% 2026-08-28
47 GPT-5.6 Sol Low OpenAI 26.0% 2026-08-27
48 Kimi K3 Max Moonshot AI 26.0% 2026-07-29
49 Qwen3.7 Max Non-reasoning Alibaba 26.0% 2026-08-27
50 Gemini 3 Flash Preview Low Google 26.0% 2026-08-05
51 Gemini 3.6 Flash Minimal Google 25.0% 2026-08-05
52 Claude Opus 4.6 Max Anthropic 25.0% 2026-07-25
53 Muse Spark 1.3 Max Meta 25.0% 2026-09-18
54 Gemini 3 Flash Preview Minimal Google 25.0% 2026-08-05
55 o3 Medium OpenAI 23.0% 2026-08-27
56 Gemini 3.6 Flash Low Google 23.0% 2026-08-05
57 GPT-5.2 High OpenAI 23.0% 2026-08-06
58 GPT-5 High OpenAI 23.0% 2026-08-05
59 Qwen3.6 35B-A3B Non-reasoning Alibaba 22.0% 2026-08-27
60 GPT-5.2 Medium OpenAI 22.0% 2026-08-28
61 Claude Opus 4.5 Thinking Anthropic 22.0% 2026-07-25
62 GPT-5.6 Luna Max OpenAI 21.0% 2026-07-28
63 Claude Opus 4.1 Thinking Anthropic 21.0% 2026-07-25
64 Qwen3.5 Flash Alibaba 20.0% 2026-08-27
65 Nemotron 3 Ultra NVIDIA 20.0% 2026-08-27
66 Mistral Large 4 High Mistral 20.0% 2026-11-02
67 GPT-5.6 Luna Non-reasoning OpenAI 20.0% 2026-08-30
68 Gemini 3 Flash Preview High Google 20.0% 2026-08-05
69 GLM-5.2 Low Zhipu AI 19.0% 2026-08-27
70 GPT-5.6 Terra Low OpenAI 19.0% 2026-08-27
71 Gemini 3.5 Flash Lite Low Google 19.0% 2026-08-05
72 GPT-5.1 Low OpenAI 19.0% 2026-08-27
73 Qwen3.6 Max Preview Non-reasoning Alibaba 19.0% 2026-08-27
74 Kimi K2.6 Moonshot AI 18.0% 2026-07-17
75 GLM-5.2 Non-reasoning Zhipu AI 18.0% 2026-08-27
76 GPT-5.5 Non-reasoning OpenAI 18.0% 2026-08-27
77 Qwen3.5 397B-A17B Non-reasoning Alibaba 18.0% 2026-08-27
78 Qwen3.6 Flash Non-reasoning Alibaba 18.0% 2026-08-27
79 GLM-5.2 Minimal Zhipu AI 18.0% 2026-08-27
80 Qwen3.7 Plus Alibaba 17.0% 2026-08-28
81 Mistral Large 4 Preview High Mistral 17.0% 2026-10-13
82 GPT-5.6 Luna Low OpenAI 17.0% 2026-08-27
83 Qwen3.7 Plus Non-reasoning Alibaba 17.0% 2026-08-27
84 GPT-5.4 Low OpenAI 17.0% 2026-08-27
85 Claude Sonnet 4.5 Thinking Anthropic 17.0% 2026-07-25
86 Qwen3.5 122B-A10B Non-reasoning Alibaba 17.0% 2026-08-27
87 Qwen3.5 Plus Non-reasoning Alibaba 17.0% 2026-08-27
88 DeepSeek V4 Pro Non-reasoning DeepSeek 17.0% 2026-08-27
89 Claude Sonnet 5 Anthropic 16.0% 2026-08-06
90 Qwen3.5 Plus Alibaba 16.0% 2026-08-05
91 GPT-5.4 Non-reasoning OpenAI 16.0% 2026-08-30
92 GPT-5.1 Medium OpenAI 16.0% 2026-08-27
93 Qwen3.5 Flash Non-reasoning Alibaba 16.0% 2026-08-27
94 GPT-5 Low OpenAI 16.0% 2026-08-27
95 Claude Sonnet 4.6 Low Anthropic 16.0% 2026-08-06
96 Claude Opus 4.6 Anthropic 15.0% 2026-08-06
97 Qwen3.7 Flash Alibaba 15.0% 2026-08-28
98 GPT-5.1 Non-reasoning OpenAI 15.0% 2026-08-27
99 GPT-5 Medium OpenAI 15.0% 2026-08-27
100 GLM-5.2 Medium Zhipu AI 15.0% 2026-08-27
101 Muse Spark 1.3 Extra High Meta 14.1% 2026-09-16
102 Claude Sonnet 4.6 Anthropic 14.0% 2026-08-06
103 GPT-5.6 Terra Non-reasoning OpenAI 14.0% 2026-08-27
104 Qwen3.7 Flash Non-reasoning Alibaba 14.0% 2026-08-27
105 GPT-5 Minimal OpenAI 14.0% 2026-08-27
106 GPT-5.2 Non-reasoning OpenAI 14.0% 2026-08-27
107 Claude Opus 4.7 Anthropic 13.0% 2026-08-06
108 GPT-4o mini OpenAI 12.0% 2026-08-27
109 GPT-4 OpenAI 12.0% 2026-08-28
110 Qwen3.6 Flash Alibaba 12.0% 2026-08-27
111 Kimi K3 Low Moonshot AI 12.0% 2026-08-29
112 Gemini 3.5 Flash Lite Minimal Google 12.0% 2026-08-05
113 Qwen3.6 Plus Non-reasoning Alibaba 12.0% 2026-08-27
114 Kimi K2.6 Non-reasoning Moonshot AI 12.0% 2026-08-29
115 GPT-5.6 Luna Medium OpenAI 12.0% 2026-08-27
116 GPT-5.4 mini Non-reasoning OpenAI 11.0% 2026-08-27
117 GPT-5 Mini Minimal OpenAI 10.0% 2026-08-27
118 GPT-5.2 Low OpenAI 10.0% 2026-08-27
119 GPT-5.6 Terra Medium OpenAI 10.0% 2026-08-28
120 Qwen3 235B-A22B Thinking 2507 Alibaba 9.0% 2026-08-27
121 GPT-5.4 nano Non-reasoning OpenAI 9.0% 2026-08-27
122 GPT-5 Nano Low OpenAI 9.0% 2026-08-27
123 GPT-5 Nano Medium OpenAI 9.0% 2026-08-27
124 GLM-5.3-Flash Max Zhipu AI 8.0% 2026-08-28
125 MiniMax-M3 Non-reasoning MiniMax 8.0% 2026-08-27
126 GPT-5 Nano High OpenAI 8.0% 2026-08-27
127 GPT-5.4 mini Low OpenAI 8.0% 2026-08-27
128 MiniMax-M3 MiniMax 7.0% 2026-08-27
129 GPT-4.1 mini OpenAI 7.0% 2026-08-27
130 GPT-6 Luna Max OpenAI 7.0% 2026-09-22
131 Qwen3.6 27B Non-reasoning Alibaba 7.0% 2026-08-27
132 o3-mini High OpenAI 7.0% 2026-08-27
133 GPT-5.4 mini Medium OpenAI 7.0% 2026-08-27
134 Claude Opus 4.6 Low Anthropic 7.0% 2026-08-06
135 Inkling Small Extra High Thinking Machines 6.0% 2026-08-15
136 GPT-5.4 nano Medium OpenAI 6.0% 2026-08-27
137 Qwen3 Max Alibaba 5.0% 2026-08-27
138 GPT-5 Nano Minimal OpenAI 5.0% 2026-08-27
139 GPT-5.4 nano High OpenAI 5.0% 2026-08-27
140 GPT-5 Mini High OpenAI 5.0% 2026-08-27
141 o4-mini High OpenAI 5.0% 2026-08-27
142 GPT-5 Mini Medium OpenAI 4.0% 2026-08-27
143 GPT-3.5-turbo OpenAI 3.0% 2026-08-27
144 GPT-5.4 nano Low OpenAI 3.0% 2026-08-27
145 GPT OSS 120B Medium OpenAI 2.0% 2026-08-27
146 GPT OSS 120B High OpenAI 0.0% 2026-08-27