Deduction

Subcategory of Reasoning. 155 models scored.

Model Leaderboard

All models ranked by their Deduction subcategory score.

# Model Deduction Reasoning Overall
1 Gemini 3.5 Flash (Reasoning) 99.50% 98.45% 94.08%
2 Gemini 3 Flash (Preview, Reasoning) 99.50% 98.05% 90.50%
3 Gemma 4 26B (Reasoning) 99.00% 95.21% 91.49%
4 Gemma 4 31B (Reasoning) 98.50% 97.19% 91.71%
5 Gemma 4 31B 98.00% 96.52% 86.91%
6 Claude Opus 4.7 97.50% 95.72% 89.93%
7 Z.AI GLM 4.6 97.50% 95.12% 89.11%
8 Gemini 2.5 Pro 97.00% 96.91% 88.53%
9 Gemma 4 26B 97.00% 88.02% 85.84%
10 Z.AI GLM 5.1 96.50% 96.83% 94.37%
11 Gemini 2.5 Flash Lite (Reasoning) 96.50% 93.86% 85.75%
12 ByteDance Seed 2.0 Lite 96.50% 95.50% 84.80%
13 Qwen3.6 Max Preview 96.00% 96.51% 94.54%
14 Gemini 3.1 Pro (Preview) 96.00% 96.01% 94.37%
15 Z.AI GLM 5 96.00% 95.89% 91.23%
16 MoonshotAI: Kimi K2.5 96.00% 95.41% 91.04%
17 GPT-5.4 Mini 96.00% 88.04% 82.43%
18 Gemini 2.5 Flash (Reasoning) 95.89% 93.81% 86.51%
19 GPT-5 Mini 95.50% 94.36% 92.62%
20 GPT-5.2 95.50% 94.54% 90.26%
21 MiniMax M2.5 95.50% 92.42% 88.71%
22 o4 Mini 95.50% 94.45% 88.35%
23 Qwen3.7 Max 95.00% 95.84% 95.75%
24 Z.AI GLM 5 Turbo 95.00% 95.67% 94.27%
25 Grok 4.3 (Reasoning) 95.00% 95.46% 93.60%
26 GPT-5.4 (Reasoning) 95.00% 94.78% 93.24%
27 Claude Opus 4.7 (Reasoning) 95.00% 95.44% 93.23%
28 GPT-5.5 (Reasoning) 95.00% 94.89% 92.98%
29 GPT-5.5 (Reasoning, Low) 95.00% 95.01% 92.59%
30 GPT-5.1 95.00% 95.14% 92.54%
31 MoonshotAI: Kimi K2.6 95.00% 95.22% 92.31%
32 GPT-5 95.00% 95.67% 91.93%
33 Qwen 3.5 397B A17B 95.00% 95.06% 91.73%
34 Qwen 3.5 122B 95.00% 94.93% 91.53%
35 GPT-5.4 (Reasoning, Low) 95.00% 94.34% 91.41%
36 Qwen 3.5 27B 95.00% 92.73% 90.85%
37 Qwen 3.6 Flash 95.00% 94.20% 90.65%
38 GPT-5.4 Mini (Reasoning) 95.00% 94.56% 90.65%
39 o4 Mini High 95.00% 95.02% 90.29%
40 GPT-5.5 95.00% 95.16% 89.09%
41 Qwen 3.6 35B 95.00% 94.08% 89.05%
42 Gemini 3 Pro (Preview) 95.00% 95.24% 88.79%
43 Z.AI GLM 4.7 95.00% 94.99% 88.69%
44 Grok 4 95.00% 96.01% 88.12%
45 Qwen 3.5 35B 95.00% 94.88% 88.00%
46 GPT-OSS 120B 95.00% 92.42% 86.44%
47 Qwen 3.5 Flash 95.00% 94.66% 86.38%
48 Grok 4 Fast 95.00% 94.89% 86.15%
49 Qwen 3.5 9B 95.00% 92.93% 86.05%
50 Gemini 3.1 Flash Lite 95.00% 92.10% 85.75%
51 GPT-5.4 Mini (Reasoning, Low) 95.00% 92.28% 85.75%
52 Gemini 3 Flash (Preview) 95.00% 94.79% 85.35%
53 Z.AI GLM 4.7 Flash 95.00% 89.50% 84.82%
54 Nemotron 3 Super 95.00% 93.11% 84.56%
55 GPT-5.4 95.00% 93.92% 84.32%
56 Inception Mercury 2 95.00% 92.03% 83.85%
57 Stealth: Aurora Alpha 95.00% 90.11% 83.79%
58 GPT-5 Nano 95.00% 89.61% 82.60%
59 Gemini 2.5 Flash Lite 95.00% 85.80% 81.08%
60 Gemini 2.5 Flash 95.00% 92.60% 80.60%
61 Inception Mercury 95.00% 85.96% 79.50%
62 GPT-5.4 Nano (Reasoning, Low) 95.00% 78.93% 79.48%
63 Gemma 3 12B 95.00% 79.42% 78.41%
64 Gemma 3 27B 95.00% 86.74% 77.85%
65 Nemotron 3 Nano 95.00% 89.91% 77.73%
66 Gemma 3 4B 95.00% 73.64% 68.57%
67 MiniMax M2.7 94.94% 93.28% 89.10%
68 Qwen 3.5 Plus (2026-04-20) 94.44% 94.99% 91.51%
69 DeepSeek V4 Pro (Reasoning) 94.44% 94.61% 90.10%
70 DeepSeek V4 Flash (Reasoning) 94.44% 95.08% 89.01%
71 Claude Sonnet 4 94.44% 94.48% 88.72%
72 Gemini 3.1 Flash Lite (Reasoning) 94.44% 91.72% 86.41%
73 Gemini 3.1 Flash Lite (Preview) 94.44% 92.15% 85.87%
74 GPT-4o Mini (temp=0) 94.44% 81.26% 78.29%
75 GPT-4.1 Nano 94.44% 70.24% 71.94%
76 Qwen 3.6 27B 93.94% 93.08% 89.72%
77 Qwen 3.5 Plus (2026-02-15) 93.94% 93.45% 85.96%
78 GPT-5.4 Nano (Reasoning) 93.89% 88.48% 81.36%
79 GPT-4o Mini (temp=1) 93.44% 80.28% 79.08%
80 GPT-5.4 Nano 93.06% 75.66% 74.40%
81 Mistral Small Creative 92.94% 87.99% 73.27%
82 Gemini 3.5 Flash (Reasoning, Minimal) 92.83% 94.12% 86.47%
83 Aion 2.0 92.78% 94.13% 89.21%
84 Z.AI GLM 4.5 92.22% 91.03% 86.27%
85 Claude Opus 4 91.94% 92.59% 87.69%
86 ByteDance Seed 1.6 Flash 91.67% 86.52% 73.27%
87 Grok 4.1 Fast 91.11% 93.58% 89.55%
88 ByteDance Seed 2.0 Mini 91.11% 92.40% 86.91%
89 Xiaomi MIMO v2.5 90.56% 92.43% 85.05%
90 Mistral Small 4 (Reasoning) 90.44% 87.78% 82.39%
91 DeepSeek V3 (2024-12-26) 90.39% 88.71% 83.68%
92 DeepSeek V3.2 90.06% 89.46% 82.25%
93 ByteDance Seed 1.6 90.00% 91.49% 90.70%
94 Stealth: Healer Alpha 90.00% 91.67% 85.93%
95 GPT-4.1 89.94% 88.46% 88.68%
96 Stealth: Hunter Alpha 89.94% 91.67% 87.34%
97 DeepSeek V3 (2025-03-24) 89.61% 88.45% 81.99%
98 Mistral Small 4 89.50% 78.72% 76.46%
99 Claude Opus 4.6 (Reasoning) 89.44% 93.77% 95.02%
100 Claude Sonnet 4.6 (Reasoning) 89.44% 92.76% 93.66%
101 Claude Opus 4.8 (Reasoning) 89.44% 93.53% 92.22%
102 Claude Opus 4.8 (Reasoning, Low) 89.44% 93.37% 92.14%
103 Claude Opus 4.5 89.44% 93.93% 89.69%
104 Claude Sonnet 4.5 89.44% 92.50% 88.03%
105 Xiaomi MIMO v2.5 Pro 89.44% 92.07% 87.36%
106 Mistral Large 3 89.44% 88.95% 85.43%
107 GPT-4o, May 13th (temp=0) 89.44% 88.58% 85.36%
108 Claude Haiku 4.5 89.44% 87.76% 85.14%
109 DeepSeek-V2 Chat 89.44% 88.70% 84.83%
110 Claude 3.5 Sonnet 89.44% 90.30% 84.24%
111 GPT-4o, May 13th (temp=1) 89.44% 85.98% 83.80%
112 Claude 3.7 Sonnet 89.44% 89.94% 83.39%
113 GPT-4.1 Mini 89.44% 85.83% 83.20%
114 GPT-4o, Aug. 6th (temp=1) 89.44% 86.91% 82.62%
115 GPT-4o, Aug. 6th (temp=0) 89.44% 87.59% 82.45%
116 Writer: Palmyra X5 89.44% 86.57% 79.57%
117 Mistral Medium 3.1 89.44% 89.32% 77.83%
118 Qwen 2.5 72B 89.44% 83.43% 75.46%
119 Ministral 3 14B 89.44% 83.24% 72.54%
120 Hermes 3 70B 89.39% 79.08% 72.57%
121 Claude Opus 4.6 88.89% 93.33% 92.35%
122 Qwen 3 32B 88.89% 86.35% 82.21%
123 Claude 3 Haiku 88.89% 77.94% 71.19%
124 Z.AI GLM 4.5 Air 88.83% 87.91% 83.12%
125 Qwen3 235B A22B Instruct 2507 88.39% 85.82% 80.10%
126 Hermes 3 405B 88.28% 85.58% 82.86%
127 MiniMax M3 87.78% 91.84% 90.88%
128 Mistral Large 2 87.78% 88.20% 82.41%
129 Mistral NeMO 87.33% 57.59% 65.04%
130 Llama 3.1 Nemotron 70B 87.22% 82.19% 74.70%
131 Llama 3.1 8B 86.78% 69.12% 63.35%
132 Grok 4.20 (Beta) 86.72% 87.05% 83.85%
133 Grok 4.3 85.89% 84.62% 78.66%
134 Llama 3.1 70B 85.00% 79.31% 78.40%
135 DeepSeek V3.1 84.94% 83.95% 82.39%
136 DeepSeek V4 Pro 84.17% 87.07% 82.63%
137 Claude Sonnet 4.6 83.89% 88.48% 91.15%
138 Mistral Small 3.2 24B 83.89% 81.71% 78.58%
139 Arcee AI: Trinity Large (Preview) 83.89% 77.24% 73.33%
140 Cydonia 24B V4.1 83.61% 76.47% 75.09%
141 Arcee AI: Trinity Mini 82.28% 76.94% 70.90%
142 Grok 4.20 81.50% 84.81% 81.70%
143 LFM2 24B 78.89% 54.88% 58.77%
144 Ministral 3 3B 77.78% 71.88% 67.22%
145 Ministral 8B 76.00% 73.78% 64.87%
146 Rocinante 12B 74.89% 54.31% 54.54%
147 Ministral 3B 72.83% 69.70% 61.29%
148 Skyfall 36B V2 70.94% 62.23% 65.76%
149 WizardLM 2 8x22b 69.94% 67.36% 71.06%
150 DeepSeek V4 Flash 68.94% 79.16% 82.02%
151 Grok 4.20 (Beta, Reasoning) 68.33% 82.64% 91.49%
152 Ministral 3 8B 68.33% 71.64% 71.76%
153 Mistral Large 66.06% 76.31% 80.15%
154 Cohere Command R+ (Aug. 2024) 63.83% 65.10% 69.03%
155 Grok 4.20 (Reasoning) 61.11% 79.11% 91.39%