Chess Puzzles

Chess tactics puzzles given as text. Run by Epoch AI.

CategoryReasoning
MetricAccuracy
SourceEpoch AI (evaluator)
LicenseCC BY 4.0
Benchmark pageepoch.ai
StatusScored
Weight0.7 within Reasoning
FitDiscrimination 1.18, noise 0.61

Results

#ModelProviderResultDate
1 Claude Fable 5.2 Max Anthropic 79.0% 2026-11-06
2 Gemini 4 Carbon High Google 78.0% 2026-11-02
3 GPT-6 Astra Pro Max OpenAI 76.0% 2026-10-15
4 GPT-6 Astra Max OpenAI 74.0% 2026-10-13
5 GPT-6.1 Sol Max OpenAI 68.0% 2026-10-13
6 GPT-5.6 Sol Pro Max OpenAI 68.0% 2026-10-14
7 Claude Opus 5.5 Max Anthropic 67.0% 2026-10-13
8 Gemini 3.7 Flash High Google 66.0% 2026-10-13
9 GPT-5.5 Extra High OpenAI 63.0% 2026-10-13
10 GPT-5.5 Pro Extra High OpenAI 62.0% 2026-10-14
11 GPT-5.2 Pro Extra High OpenAI 60.0% 2026-10-14
12 Gemini 4 Argon High Google 59.0% 2026-10-13
13 Claude Sonnet 5.5 Max Anthropic 59.0% 2026-10-13
14 GPT-5.6 Sol Max OpenAI 59.0% 2026-10-13
15 GPT-5.4 Pro Extra High OpenAI 59.0% 2026-10-14
16 Gemini 3.8 Flash High Google 58.0% 2026-10-13
17 GPT-6 Sol Max OpenAI 56.0% 2026-10-13
18 Claude Haiku 5.5 Max Anthropic 55.0% 2026-10-13
19 Claude Opus 4.8 Max Anthropic 52.0% 2026-10-13
20 Gemini 3.1 Pro Preview High Google 51.0% 2026-10-13
21 DeepSeek V4.1 Flash Max DeepSeek 48.0% 2026-10-13
22 GPT-5.4 Extra High OpenAI 48.0% 2026-10-13
23 GPT-5.2 Extra High OpenAI 48.0% 2026-10-14
24 Claude Fable 5.1 Max Anthropic 48.0% 2026-10-13
25 GPT-5 Pro High OpenAI 48.0% 2026-10-14
26 Claude Fable 5 Max Anthropic 47.0% 2026-10-13
27 GPT-5.6 Terra Max OpenAI 47.0% 2026-10-13
28 DeepSeek V4.1 Flash 1017 Max DeepSeek 46.0% 2026-10-19
29 DeepSeek V4 Pro 0813 Max DeepSeek 46.0% 2026-10-13
30 Qwen3.8 Max Extra High Alibaba 46.0% 2026-10-13
31 Kimi K3.1 Pro Max Moonshot AI 45.0% 2026-11-06
32 Claude Sonnet 5 Max Anthropic 45.0% 2026-10-13
33 Claude Fable 5 High Anthropic 45.0% 2026-10-13
34 Gemini 3.5 Flash High Google 45.0% 2026-10-13
35 GPT-5.2 High OpenAI 45.0% 2026-10-13
36 GPT-5.4 nano Extra High OpenAI 45.0% 2026-10-14
37 Claude Opus 5 Max Anthropic 43.0% 2026-10-13
38 GPT-5.6 Luna Max OpenAI 43.0% 2026-10-13
39 Gemini 3.6 Flash Low Google 42.0% 2026-10-13
40 Gemini 3.5 Flash Low Google 42.0% 2026-10-13
41 GPT-5.4 High OpenAI 42.0% 2026-10-13
42 GPT-5.1 High OpenAI 42.0% 2026-10-13
43 Grok 4.7 Extra High xAI 41.0% 2026-10-13
44 Claude Opus 4.7 Max Anthropic 41.0% 2026-10-13
45 Doubao Seed Evolving (Nov 2026) High ByteDance 40.0% 2026-11-02
46 Gemini 3 Flash Preview High Google 40.0% 2026-10-13
47 Gemini 3.6 Flash High Google 40.0% 2026-10-13
48 Qwen3.8 Max 0902 Extra High Alibaba 37.0% 2026-10-13
49 GLM-5.3 Max Zhipu AI 37.0% 2026-10-13
50 Kimi K3 High Moonshot AI 37.0% 2026-10-13
51 o3 Medium OpenAI 37.0% 2026-10-13
52 GPT-5.4 Medium OpenAI 37.0% 2026-10-13
53 Muse Spark 1.3 Max Meta 37.0% 2026-10-13
54 GPT-6 Luna Max OpenAI 36.0% 2026-10-13
55 Kimi K3 Max Moonshot AI 36.0% 2026-10-13
56 o3 Low OpenAI 36.0% 2026-10-13
57 Claude Opus 4.7 Extra High Anthropic 36.0% 2026-10-13
58 GPT-5.4 nano High OpenAI 36.0% 2026-10-13
59 GPT-5.2 Medium OpenAI 36.0% 2026-10-13
60 GPT-5 High OpenAI 36.0% 2026-10-13
61 o3 High OpenAI 36.0% 2026-10-13
62 Grok 4.5 High xAI 35.0% 2026-10-13
63 Claude Sonnet 5 Extra High Anthropic 34.0% 2026-10-13
64 Muse Spark 1.3 Extra High Meta 34.0% 2026-10-13
65 Muse Spark 1.1 Extra High Meta 34.0% 2026-10-13
66 Mistral Large 4 Preview High Mistral 33.0% 2026-10-15
67 Grok 4.6 Extra High xAI 33.0% 2026-10-13
68 Grok 4.6 High xAI 33.0% 2026-10-13
69 DeepSeek V4 Flash 0731 Max DeepSeek 32.0% 2026-10-13
70 GPT-5 Medium OpenAI 32.0% 2026-10-13
71 GPT-5.6 Sol Low OpenAI 31.0% 2026-10-13
72 Claude Fable 5 Low Anthropic 31.0% 2026-10-13
73 GPT-5.5 Low OpenAI 30.0% 2026-10-13
74 Grok 4.20 (Reasoning) xAI 29.0% 2026-10-13
75 Mistral Large 4 High Mistral 29.0% 2026-11-02
76 Gemini 3.6 Flash Minimal Google 29.0% 2026-10-13
77 Claude Opus 4.8 Low Anthropic 29.0% 2026-10-13
78 Qwen3.7 Max Alibaba 27.0% 2026-10-13
79 GPT-5.6 Terra Low OpenAI 27.0% 2026-10-13
80 Claude Opus 5 Low Anthropic 27.0% 2026-10-13
81 Gemini 3.5 Flash Minimal Google 27.0% 2026-10-13
82 GPT-5 Low OpenAI 27.0% 2026-10-13
83 Muse Spark 1.2 Extra High Meta 27.0% 2026-10-13
84 Doubao Seed 2.1 Turbo High ByteDance 26.0% 2026-10-30
85 GPT-5.2 Low OpenAI 26.0% 2026-10-13
86 Claude Opus 4.7 Low Anthropic 26.0% 2026-10-13
87 GPT-5 Mini High OpenAI 24.0% 2026-10-13
88 Claude Haiku 5.5 Low Anthropic 24.0% 2026-10-13
89 Claude Sonnet 4.6 Max Anthropic 23.0% 2026-10-13
90 GPT-5.4 mini High OpenAI 23.0% 2026-10-13
91 GPT-5.1 Low OpenAI 23.0% 2026-10-13
92 o4-mini Medium OpenAI 22.0% 2026-10-13
93 GLM-5.2 Max Zhipu AI 22.0% 2026-10-14
94 GPT-5 Nano High OpenAI 22.0% 2026-10-13
95 GLM-4.7 Zhipu AI 21.0% 2026-10-14
96 Doubao Seed 2.1 Pro High ByteDance 21.0% 2026-10-28
97 GLM-5.3-Flash Max Zhipu AI 21.0% 2026-10-13
98 Claude Opus 4.6 Max Anthropic 21.0% 2026-10-13
99 Gemini 3.5 Flash Lite Low Google 21.0% 2026-10-13
100 Inkling Extra High Thinking Machines 21.0% 2026-10-13
101 Kimi K2.6 Moonshot AI 20.0% 2026-10-13
102 GLM-5.1 Zhipu AI 20.0% 2026-10-14
103 Doubao Seed 2.0 Lite (Apr 2026) High ByteDance 20.0% 2026-10-28
104 Doubao-Seed-2.0 Pro High ByteDance 20.0% 2026-10-28
105 GPT-5.4 nano Low OpenAI 20.0% 2026-10-13
106 GLM-5.2 Non-reasoning Zhipu AI 19.0% 2026-10-14
107 Claude Opus 4.8 Non-reasoning Anthropic 19.0% 2026-10-13
108 GPT-5.4 Low OpenAI 19.0% 2026-10-13
109 Claude Sonnet 4.6 High Anthropic 19.0% 2026-10-13
110 DeepSeek V4 Pro Max DeepSeek 19.0% 2026-10-13
111 GLM-5.2 Low Zhipu AI 18.0% 2026-10-14
112 Kimi K3 Low Moonshot AI 18.0% 2026-10-13
113 Grok 4.3 High xAI 18.0% 2026-10-13
114 MiniMax-M3 MiniMax 17.0% 2026-10-13
115 GPT-5.4 mini Extra High OpenAI 17.0% 2026-10-13
116 GPT-5.6 Luna Low OpenAI 17.0% 2026-10-13
117 Qwen3.6 Plus Alibaba 16.0% 2026-10-13
118 Gemini 3.5 Flash Lite High Google 16.0% 2026-10-13
119 Claude Sonnet 4.6 Medium Anthropic 16.0% 2026-10-13
120 GPT-5 Mini Low OpenAI 16.0% 2026-10-13
121 DeepSeek V4 Pro High DeepSeek 16.0% 2026-10-13
122 GPT-5 Nano Low OpenAI 15.0% 2026-10-13
123 o4-mini Low OpenAI 15.0% 2026-10-13
124 o3-mini High OpenAI 15.0% 2026-10-13
125 Kimi K2.7 Code Moonshot AI 14.0% 2026-10-13
126 Qwen3.5 Flash Alibaba 14.0% 2026-10-13
127 Claude Opus 4.5 Thinking Anthropic 14.0% 2026-10-13
128 o4-mini High OpenAI 13.0% 2026-10-13
129 Qwen3.6 Flash Alibaba 12.0% 2026-10-13
130 GPT-4.1 OpenAI 11.0% 2026-10-13
131 o1 Low OpenAI 11.0% 2026-10-13
132 GPT-5 Minimal OpenAI 11.0% 2026-10-13
133 o3-mini Medium OpenAI 11.0% 2026-10-13
134 Claude Sonnet 4.5 Thinking Anthropic 10.0% 2026-10-13
135 o1 High OpenAI 10.0% 2026-10-13
136 GPT-4o (2024-08-06) OpenAI 9.0% 2026-10-13
137 Qwen3.5 Plus Alibaba 9.0% 2026-10-13
138 GPT-5.5 Non-reasoning OpenAI 9.0% 2026-10-13
139 MiniMax-M3 Non-reasoning MiniMax 8.0% 2026-10-13
140 Claude Haiku 4.5 Thinking Anthropic 8.0% 2026-10-13
141 o1 Medium OpenAI 8.0% 2026-10-13
142 GPT-5.6 Sol Non-reasoning OpenAI 7.0% 2026-10-13
143 Gemini 3.5 Flash Lite Minimal Google 7.0% 2026-10-13
144 Claude Sonnet 4.5 Anthropic 6.0% 2026-10-13
145 GPT-5.1 Non-reasoning OpenAI 6.0% 2026-10-13
146 o3-mini Low OpenAI 6.0% 2026-10-13
147 Claude Opus 4.5 Anthropic 5.0% 2026-10-13
148 Seed 1.6 ByteDance 5.0% 2026-10-30
149 GPT-5.6 Terra Non-reasoning OpenAI 5.0% 2026-10-13
150 Claude 3 Opus Anthropic 4.0% 2026-10-13
151 DeepSeek V4 Pro Non-reasoning DeepSeek 4.0% 2026-10-13
152 GPT-4o mini OpenAI 3.0% 2026-10-13
153 GPT-5.6 Luna Non-reasoning OpenAI 3.0% 2026-10-13
154 GPT-5.4 Non-reasoning OpenAI 3.0% 2026-10-13
155 GPT-5.4 mini Non-reasoning OpenAI 2.0% 2026-10-13
156 GPT-5.4 nano Non-reasoning OpenAI 2.0% 2026-10-13
157 GPT-5 Nano Minimal OpenAI 2.0% 2026-10-13
158 GPT-5.2 Non-reasoning OpenAI 2.0% 2026-10-13
159 GPT-4.1 mini OpenAI 1.0% 2026-10-13
160 GPT-4.1 nano OpenAI 0.0% 2026-10-13
161 GPT-5 Mini Minimal OpenAI 0.0% 2026-10-13