ARC-AGI-1

The original ARC puzzles. Near saturation.

CategoryReasoning
MetricAccuracy
SourceEpoch AI (evaluator)
LicenseCC BY 4.0
Benchmark pagearcprize.org
StatusScored
Weight0.6 within Reasoning
FitDiscrimination 2.09, noise 0.46

Results

#ModelProviderResultDate
1 Gemini 4 Argon High Google 98.5% 2026-11-11
2 Gemini 3.8 Flash High Google 98.5% 2026-11-11
3 Claude Fable 5 Max Anthropic 98.5% 2026-11-11
4 GPT-6 Astra High OpenAI 98.5% 2026-11-11
5 GPT-6 Astra Extra High OpenAI 98.5% 2026-11-11
6 Claude Fable 5 Extra High Anthropic 98.5% 2026-11-11
7 Claude Opus 5.5 High Anthropic 98.5% 2026-11-11
8 Gemini 4 Argon Medium Google 98.5% 2026-11-11
9 GPT-6.1 Sol High OpenAI 98.5% 2026-11-11
10 GPT-6.1 Sol Extra High OpenAI 98.5% 2026-11-11
11 Claude Fable 5.2 Extra High Anthropic 98.5% 2026-11-11
12 Gemini 4 Carbon Medium Google 98.5% 2026-11-11
13 Gemini 3.1 Pro Preview Google 98.0% 2026-11-11
14 Claude Fable 5.2 Max Anthropic 97.5% 2026-11-11
15 Gemini 4 Carbon High Google 97.5% 2026-11-11
16 Claude Opus 5.5 Max Anthropic 97.5% 2026-11-11
17 GPT-6 Astra Max OpenAI 97.5% 2026-11-11
18 Claude Opus 5 Max Anthropic 97.5% 2026-11-11
19 Claude Fable 5.1 Max Anthropic 97.5% 2026-11-11
20 GPT-6 Astra Medium OpenAI 97.5% 2026-11-11
21 Claude Opus 5 High Anthropic 97.5% 2026-11-11
22 GPT-5.6 Sol Extra High OpenAI 97.5% 2026-11-11
23 Gemini 3.8 Flash Medium Google 97.5% 2026-11-11
24 Claude Opus 5.5 Extra High Anthropic 97.5% 2026-11-11
25 Claude Opus 5.5 Medium Anthropic 97.5% 2026-11-11
26 Claude Fable 5.2 High Anthropic 97.5% 2026-11-11
27 Claude Fable 5.2 Medium Anthropic 97.5% 2026-11-11
28 GPT-5.6 Sol High OpenAI 97.0% 2026-11-11
29 GPT-6.1 Sol Max OpenAI 96.5% 2026-11-11
30 Claude Sonnet 5.5 Max Anthropic 96.5% 2026-11-11
31 GPT-5.6 Terra Max OpenAI 96.5% 2026-11-11
32 GPT-5.6 Sol Max OpenAI 96.5% 2026-11-11
33 GPT-5.5 Pro High OpenAI 96.5% 2026-11-11
34 GPT-6 Astra Low OpenAI 96.5% 2026-11-11
35 Claude Fable 5.1 Extra High Anthropic 96.5% 2026-11-11
36 Claude Sonnet 5.5 Extra High Anthropic 96.5% 2026-11-11
37 Inkling 2 Small Extra High Thinking Machines 96.5% 2026-11-11
38 Gemini 3 Deep Think Google 96.0% 2026-11-11
39 Claude Fable 5.1 High Anthropic 96.0% 2026-11-11
40 GPT-6 Sol Max OpenAI 95.5% 2026-11-11
41 Gemini 3.7 Flash High Google 95.5% 2026-11-11
42 Claude Fable 5 High Anthropic 95.5% 2026-11-11
43 GPT-6.1 Sol Medium OpenAI 95.5% 2026-11-11
44 GPT-5.5 Pro Extra High OpenAI 95.0% 2026-11-11
45 GPT-5.5 Extra High OpenAI 95.0% 2026-11-11
46 DeepSeek V4.1 Flash Max DeepSeek 94.5% 2026-11-11
47 Kimi K3 Max Moonshot AI 94.5% 2026-11-11
48 GPT-5.4 Pro Extra High OpenAI 94.5% 2026-11-11
49 GPT-5.5 High OpenAI 94.5% 2026-11-11
50 Claude Fable 5.1 Medium Anthropic 94.5% 2026-11-11
51 DeepSeek V4.1 Flash 1017 High DeepSeek 94.5% 2026-11-11
52 Claude Fable 5.2 Low Anthropic 94.5% 2026-11-11
53 Claude Opus 4.6 Thinking Anthropic 94.0% 2026-11-11
54 GPT-5.6 Terra Extra High OpenAI 94.0% 2026-11-11
55 GPT-5.4 Extra High OpenAI 93.7% 2026-11-11
56 DeepSeek V4.1 Flash 1017 Max DeepSeek 93.5% 2026-11-11
57 Claude Opus 4.7 High Anthropic 93.5% 2026-11-11
58 GPT-6.1 Sol Low OpenAI 93.5% 2026-11-11
59 GPT-5.4 High OpenAI 92.7% 2026-11-11
60 GPT-6 Sol Extra High OpenAI 92.7% 2026-11-11
61 Claude Opus 4.8 Max Anthropic 92.5% 2026-11-11
62 Gemini 3.5 Flash High Google 92.5% 2026-11-11
63 GPT-5.6 Sol Medium OpenAI 92.5% 2026-11-11
64 Claude Fable 5 Medium Anthropic 92.5% 2026-11-11
65 GPT-5.5 Medium OpenAI 92.2% 2026-11-11
66 Claude Opus 4.7 Max Anthropic 92.0% 2026-11-11
67 GPT-5.6 Terra High OpenAI 92.0% 2026-11-11
68 Claude Sonnet 5.5 High Anthropic 92.0% 2026-11-11
69 Claude Opus 4.8 High Anthropic 92.0% 2026-11-11
70 DeepSeek V4.1 Flash 1017 Low DeepSeek 91.5% 2026-11-11
71 Claude Opus 4.8 Medium Anthropic 91.5% 2026-11-11
72 Gemini 3.6 Flash High Google 91.2% 2026-11-11
73 Gemini 3.7 Flash Medium Google 91.2% 2026-11-11
74 GLM-5.3-Flash Max Zhipu AI 91.0% 2026-11-11
75 Claude Opus 4.7 Low Anthropic 91.0% 2026-11-11
76 GPT-6 Sol High OpenAI 91.0% 2026-11-11
77 Claude Opus 4.7 Medium Anthropic 91.0% 2026-11-11
78 Claude Fable 5 Low Anthropic 90.5% 2026-11-11
79 GPT-5.2 Pro Extra High OpenAI 90.5% 2026-11-11
80 Gemini 3.8 Flash Low Google 90.5% 2026-11-11
81 DeepSeek V4.1 Flash Low DeepSeek 90.5% 2026-11-11
82 DeepSeek V4 Pro 0813 Low DeepSeek 90.5% 2026-11-11
83 Grok 4.7 High xAI 90.2% 2026-11-11
84 DeepSeek V4 Pro 0813 Max DeepSeek 90.0% 2026-11-11
85 Claude Fable 5.1 Low Anthropic 90.0% 2026-11-11
86 Grok 4.7 Medium xAI 90.0% 2026-11-11
87 Grok 4.20 (Reasoning) xAI 89.5% 2026-11-11
88 Grok 4.7 Extra High xAI 89.5% 2026-11-11
89 Muse Spark Open Extra High Meta 89.0% 2026-11-11
90 Claude Haiku 5.5 Max Anthropic 89.0% 2026-11-11
91 DeepSeek V4 Flash 0731 Max DeepSeek 89.0% 2026-11-11
92 Claude Opus 5.5 Low Anthropic 88.5% 2026-11-11
93 DeepSeek V4.1 Flash High DeepSeek 88.5% 2026-11-11
94 Muse Spark Open High Meta 88.5% 2026-11-11
95 Claude Opus 4.8 Low Anthropic 88.0% 2026-11-11
96 GPT-5.6 Luna Max OpenAI 88.0% 2026-11-11
97 GPT-5.6 Luna Extra High OpenAI 87.7% 2026-11-11
98 Claude Haiku 5.5 Extra High Anthropic 87.5% 2026-11-11
99 Grok 4.6 Medium xAI 87.5% 2026-11-11
100 Qwen3.8 27B Extra High Alibaba 87.5% 2026-11-11
101 DeepSeek V4 Pro 0813 High DeepSeek 87.2% 2026-11-11
102 Grok 4.5 Medium xAI 87.2% 2026-11-11
103 Grok 4.6 Extra High xAI 87.0% 2026-11-11
104 Grok 4.6 High xAI 87.0% 2026-11-11
105 DeepSeek V4 Flash 0731 High DeepSeek 87.0% 2026-11-11
106 GPT-6 Luna Max OpenAI 86.7% 2026-11-11
107 Kimi K3 High Moonshot AI 86.7% 2026-11-11
108 Claude Sonnet 4.6 High Anthropic 86.5% 2026-11-11
109 GPT-5.2 Extra High OpenAI 86.2% 2026-11-11
110 GPT-5.4 Medium OpenAI 86.2% 2026-11-11
111 Claude Sonnet 4.6 Max Anthropic 86.0% 2026-11-11
112 GPT-6 Astra Non-reasoning OpenAI 86.0% 2026-11-11
113 Grok 4.5 High xAI 85.7% 2026-11-11
114 GPT-5.2 Pro High OpenAI 85.7% 2026-11-11
115 Gemini 4 Carbon Low Google 85.5% 2026-11-11
116 Gemini 3.7 Flash Low Google 85.2% 2026-11-11
117 Gemini 3 Flash Preview Google 84.7% 2026-11-11
118 Inkling Small Extra High Thinking Machines 84.0% 2026-11-11
119 DeepSeek V4 Flash 0731 Low DeepSeek 84.0% 2026-11-11
120 GPT-6 Sol Medium OpenAI 83.7% 2026-11-11
121 Gemini 3.6 Flash Medium Google 83.2% 2026-11-11
122 Claude Haiku 5.5 High Anthropic 81.5% 2026-11-11
123 GPT-5.2 Pro Medium OpenAI 81.2% 2026-11-11
124 Claude Opus 4.5 Thinking Anthropic 80.0% 2026-11-11
125 Inkling Thinking Machines 79.5% 2026-11-11
126 Grok 4.5 Low xAI 79.2% 2026-11-11
127 GPT-5.2 High OpenAI 78.7% 2026-11-11
128 Inkling Small High Thinking Machines 78.0% 2026-11-11
129 Muse Spark Open Medium Meta 77.7% 2026-11-11
130 GLM-5.2 Zhipu AI 77.0% 2026-11-11
131 GPT-5.6 Terra Medium OpenAI 77.0% 2026-11-11
132 Gemini 3.6 Flash Low Google 76.5% 2026-11-11
133 GPT-5.6 Luna High OpenAI 76.5% 2026-11-11
134 Gemini 4 Argon Low Google 76.5% 2026-11-11
135 GPT-5.5 Low OpenAI 76.2% 2026-11-11
136 Gemini 3 Pro Preview Google 75.0% 2026-11-11
137 Grok 4.6 Low xAI 74.8% 2026-11-11
138 GPT-5.6 Sol Low OpenAI 74.5% 2026-11-11
139 Beam 1.0 Extra High Reflection 74.0% 2026-11-11
140 GPT-6 Luna Extra High OpenAI 73.0% 2026-11-11
141 GPT-5.1 High OpenAI 72.8% 2026-11-11
142 GPT-5.2 Medium OpenAI 72.7% 2026-11-11
143 GPT-6 Sol Low OpenAI 72.2% 2026-11-11
144 GLM-5.3-Flash High Zhipu AI 71.8% 2026-11-11
145 GPT-6 Luna High OpenAI 70.3% 2026-11-11
146 GPT-5 Pro High OpenAI 70.2% 2026-11-11
147 GPT-5 Pro OpenAI 70.2% 2026-11-11
148 Claude Sonnet 5.5 Medium Anthropic 70.0% 2026-11-11
149 Qwen3.8 27B Low Alibaba 69.2% 2026-11-11
150 Qwen3.8 27B Medium Alibaba 68.7% 2026-11-11
151 GPT-5.4 Low OpenAI 68.2% 2026-11-11
152 Beam 1.0 High Reflection 67.7% 2026-11-11
153 Claude Sonnet 5.5 Low Anthropic 67.0% 2026-11-11
154 Inkling Small Medium Thinking Machines 67.0% 2026-11-11
155 Grok 4 xAI 66.7% 2026-11-11
156 Kimi K3 Low Moonshot AI 65.7% 2026-11-11
157 GPT-5 High OpenAI 65.7% 2026-11-11
158 Kimi K2.5 Moonshot AI 65.3% 2026-11-11
159 MiniMax-M2.5 MiniMax 63.7% 2026-11-11
160 GPT-5.4 mini Extra High OpenAI 63.7% 2026-11-11
161 Claude Sonnet 4.5 Thinking Anthropic 63.7% 2026-11-11
162 Grok 4.7 Low xAI 63.3% 2026-11-11
163 GPT-6 Luna Medium OpenAI 61.0% 2026-11-11
164 o3 High OpenAI 60.8% 2026-11-11
165 GPT-5.6 Terra Low OpenAI 60.2% 2026-11-11
166 o3-pro High OpenAI 59.3% 2026-11-11
167 o4-mini High OpenAI 58.7% 2026-11-11
168 Muse Spark Open Low Meta 58.5% 2026-11-11
169 GPT-5.4 mini High OpenAI 58.0% 2026-11-11
170 GPT-5.1 Medium OpenAI 57.7% 2026-11-11
171 Claude Haiku 5.5 Medium Anthropic 57.3% 2026-11-11
172 DeepSeek V3.2 DeepSeek 57.0% 2026-11-11
173 o3-pro Medium OpenAI 57.0% 2026-11-11
174 GPT-5.6 Luna Medium OpenAI 56.5% 2026-11-11
175 GPT-5 Medium OpenAI 56.2% 2026-11-11
176 GPT-5.2 Low OpenAI 55.7% 2026-11-11
177 Beam 1.0 Medium Reflection 55.7% 2026-11-11
178 GPT-5 Mini High OpenAI 54.3% 2026-11-11
179 o3 Medium OpenAI 53.8% 2026-11-11
180 Gemini 3.5 Flash Lite High Google 53.5% 2026-11-11
181 Inkling Small Low Thinking Machines 52.5% 2026-11-11
182 GPT-5.4 nano Extra High OpenAI 51.5% 2026-11-11
183 Claude Haiku 5.5 Low Anthropic 50.3% 2026-11-11
184 Gemini 3.5 Flash Minimal Google 48.8% 2026-11-11
185 Grok 4 Fast xAI 48.5% 2026-11-11
186 Inkling Small Minimal Thinking Machines 48.5% 2026-11-11
187 Claude Haiku 4.5 Thinking Anthropic 47.7% 2026-11-11
188 GLM-5.3-Flash Low Zhipu AI 47.0% 2026-11-11
189 Beam 1.0 Low Reflection 45.0% 2026-11-11
190 GLM-5 Zhipu AI 44.7% 2026-11-11
191 o3-pro Low OpenAI 44.3% 2026-11-11
192 GPT-5 Low OpenAI 44.0% 2026-11-11
193 o4-mini Medium OpenAI 41.8% 2026-11-11
194 o3 Low OpenAI 41.5% 2026-11-11
195 Gemini 2.5 Pro Thinking Google 41.0% 2026-11-11
196 GPT-5.4 mini Medium OpenAI 40.8% 2026-11-11
197 Claude Opus 4.5 Anthropic 40.0% 2026-11-11
198 tiny-recursion-model Unknown 40.0% 2026-11-11
199 Claude Sonnet 4 Thinking Anthropic 40.0% 2026-11-11
200 GPT-5.4 nano High OpenAI 38.2% 2026-11-11
201 GPT-6 Luna Low OpenAI 37.7% 2026-11-11
202 GPT-5 Mini Medium OpenAI 37.3% 2026-11-11
203 Claude Opus 4 Thinking Anthropic 35.7% 2026-11-11
204 Gemini 3.6 Flash Minimal Google 34.5% 2026-11-11
205 o3-mini High OpenAI 34.5% 2026-11-11
206 GPT-5.6 Luna Low OpenAI 34.2% 2026-11-11
207 Qwen3.8 27B Non-reasoning Alibaba 34.0% 2026-11-11
208 Gemini 2.5 Flash Preview 05-20 Google 33.3% 2026-11-11
209 Gemini 2.5 Flash Preview 05-20 Thinking Google 33.3% 2026-11-11
210 GPT-5.1 Low OpenAI 33.2% 2026-11-11
211 Gemini 2.5 Pro Preview 06-05 Google 33.0% 2026-11-11
212 GPT-5.4 nano Medium OpenAI 33.0% 2026-11-11
213 Gemini 3.5 Flash Lite Medium Google 32.3% 2026-11-11
214 Gemini 2.5 Flash (Apr 2025) Google 32.3% 2026-11-11
215 Gemini 2.5 Pro Preview 06-05 Thinking Google 31.3% 2026-11-11
216 o1 Medium OpenAI 30.7% 2026-11-11
217 GPT-6 Sol Non-reasoning OpenAI 29.3% 2026-11-11
218 Claude Sonnet 3.7 Thinking Anthropic 28.6% 2026-11-11
219 codex-mini-2025-05-16 Unknown 27.3% 2026-11-11
220 o1 Low OpenAI 27.2% 2026-11-11
221 GPT-5 Mini Low OpenAI 26.3% 2026-11-11
222 Claude Sonnet 4.5 Anthropic 25.5% 2026-11-11
223 Claude Sonnet 4 Anthropic 23.8% 2026-11-11
224 o1-pro Low OpenAI 23.3% 2026-11-11
225 Claude Opus 4 Anthropic 22.5% 2026-11-11
226 o3-mini Medium OpenAI 22.3% 2026-11-11
227 o4-mini Low OpenAI 21.3% 2026-11-11
228 DeepSeek R1 0528 DeepSeek 21.2% 2026-11-11
229 GPT-5 Nano Medium OpenAI 20.7% 2026-11-11
230 GPT-5.4 nano Low OpenAI 18.3% 2026-11-11
231 o1-preview OpenAI 18.0% 2026-11-11
232 Gemini 3.5 Flash Lite Low Google 17.0% 2026-11-11
233 GPT-5 Nano High OpenAI 16.7% 2026-11-11
234 Grok-3 mini Low xAI 16.5% 2026-11-11
235 DeepSeek-R1 DeepSeek 15.8% 2026-11-11
236 o3-mini Low OpenAI 14.5% 2026-11-11
237 Claude Haiku 4.5 Anthropic 14.3% 2026-11-11
238 o1-mini OpenAI 14.0% 2026-11-11
239 o1-mini Medium OpenAI 14.0% 2026-11-11
240 Claude Sonnet 3.7 Anthropic 13.6% 2026-11-11
241 DeepSeek V4 Pro 0813 Non-reasoning DeepSeek 13.0% 2026-11-11
242 GPT-5.4 mini Low OpenAI 13.0% 2026-11-11
243 GPT-5.2 OpenAI 12.3% 2026-11-11
244 GPT-5.2 Non-reasoning OpenAI 12.3% 2026-11-11
245 DeepSeek V4 Flash 0731 Non-reasoning DeepSeek 11.8% 2026-11-11
246 Qwen3 235B-A22B Instruct 2507 Alibaba 11.0% 2026-11-11
247 GPT-4.5 OpenAI 10.3% 2026-11-11
248 GPT-6 Luna Non-reasoning OpenAI 8.8% 2026-11-11
249 DeepSeek V4.1 Flash Non-reasoning DeepSeek 7.8% 2026-11-11
250 Gemini 3.5 Flash Lite Minimal Google 7.5% 2026-11-11
251 Magistral Medium 1.1 Mistral 6.1% 2026-11-11
252 GPT-5 Minimal OpenAI 6.0% 2026-11-11
253 GPT-5.1 Non-reasoning OpenAI 5.8% 2026-11-11
254 DeepSeek V4.1 Flash 1017 Non-reasoning DeepSeek 5.7% 2026-11-11
255 GPT-4.1 OpenAI 5.5% 2026-11-11
256 Grok 3 xAI 5.5% 2026-11-11
257 Inkling Small Non-reasoning Thinking Machines 5.5% 2026-11-11
258 GPT-5 Mini Minimal OpenAI 5.3% 2026-11-11
259 Magistral Small Mistral 5.0% 2026-11-11
260 GPT-4o (2024-11-20) OpenAI 4.5% 2026-11-11
261 Llama 4 Maverick Meta 4.4% 2026-11-11
262 GPT-5 Nano Low OpenAI 4.0% 2026-11-11
263 GPT-4.1 mini OpenAI 3.5% 2026-11-11
264 GPT-5 Nano Minimal OpenAI 1.5% 2026-11-11
265 Llama 4 Scout Meta 0.5% 2026-11-11
266 GPT-4.1 nano OpenAI 0.0% 2026-11-11