Skip to content

Benchmark

OTIS Mock AIME

Math benchmarks measure how well models solve maths problems. Scores are percentages.

Models measured
196
Scores recorded
302
Sources
1
IQ weight
1.62

Score distribution

62.7%
Mean
71.1%
Median
32.9%
Std dev
6.5%
10th percentile
98.3%
90th percentile
0%
Min
0%100%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.62 for chat models, measured over 196 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Qwen 3 235B A22B Thinking 2507· Epoch AI Benchmarking Hub data86.7%
102Qwen3.5 Plus· Epoch AI Benchmarking Hub data86.7%
103Qwen3.6 35B A3B· Epoch AI Benchmarking Hub data86.7%
104Qwen3.7 Flash· Epoch AI Benchmarking Hub data86.7%
105GLM 5.2max· Epoch AI Benchmarking Hub data86.4%
106Claude Opus 4.532k· Epoch AI Benchmarking Hub data86.1%
107Claude Sonnet 4.632k· Epoch AI Benchmarking Hub data85.8%
108GPT 5.1medium· Epoch AI Benchmarking Hub data85.6%
109Claude Opus 4.8none· Epoch AI Benchmarking Hub data84.4%
110GPT 5.4low· Epoch AI Benchmarking Hub data84.4%
111GPT 5.5low· Epoch AI Benchmarking Hub data84.4%
112OpenAI o3medium· Epoch AI Benchmarking Hub data84.4%
113Qwen3.5 Flash· Epoch AI Benchmarking Hub data84.4%
114Qwen3.6 Flash· Epoch AI Benchmarking Hub data84.4%
115Gemini 2.5 Pro· Epoch AI Benchmarking Hub data84.2%
116Grok 4· Epoch AI Benchmarking Hub data84%
117OpenAI o3high· Epoch AI Benchmarking Hub data83.9%
118GLM 4.7· Epoch AI Benchmarking Hub data83.3%
119Kimi K2 Thinking· Epoch AI Benchmarking Hub data83.1%
120Claude Sonnet 4.6medium· Epoch AI Benchmarking Hub data82.2%
121Gemini 3.6 Flashlow· Epoch AI Benchmarking Hub data82.2%
122Gemma 4 26B A4Bminimal· Epoch AI Benchmarking Hub data82.2%
123Qwen3.5 397B A17Bnone· Epoch AI Benchmarking Hub data82.2%
124Claude Opus 4.516k· Epoch AI Benchmarking Hub data81.7%
125o4 Minihigh· Epoch AI Benchmarking Hub data81.7%
126GPT-5 nanohigh· Epoch AI Benchmarking Hub data81.1%
127Claude Sonnet 5max· Epoch AI Benchmarking Hub data80%
128Gemini 3.1 Flash Litehigh· Epoch AI Benchmarking Hub data80%
129Gemini 3.5 Flashminimal· Epoch AI Benchmarking Hub data80%
130Gemini 3.6 Flashminimal· Epoch AI Benchmarking Hub data80%
131GLM 5· Epoch AI Benchmarking Hub data80%
132Qwen3.7 Plusnone· Epoch AI Benchmarking Hub data80%
133GPT 5.2low· Epoch AI Benchmarking Hub data78.9%
134GPT-5 minimedium· Epoch AI Benchmarking Hub data78.3%
135Claude Sonnet 4.532k· Epoch AI Benchmarking Hub data77.8%
136Claude Sonnet 4.559k· Epoch AI Benchmarking Hub data77.8%
137Grok 3 Minihigh, beta· Epoch AI Benchmarking Hub data77.8%
138Qwen3.7 Flashnone· Epoch AI Benchmarking Hub data77.8%
139o3 Minihigh· Epoch AI Benchmarking Hub data76.9%
140Claude Sonnet 4.6high· Epoch AI Benchmarking Hub data75.6%
141GLM 5.2low· Epoch AI Benchmarking Hub data75.6%
142GPT-5 nanomedium· Epoch AI Benchmarking Hub data74.2%
143Gemma 4 31B ITminimal· Epoch AI Benchmarking Hub data73.3%
144o1high· Epoch AI Benchmarking Hub data73.3%
145o1medium· Epoch AI Benchmarking Hub data73.3%
146o4 Minimedium· Epoch AI Benchmarking Hub data73.3%
147Qwen3 Max· Epoch AI Benchmarking Hub data73.3%
148gemini-2.5-flash-preview-04-17· Epoch AI Benchmarking Hub data73.1%
149Claude Sonnet 432k· Epoch AI Benchmarking Hub data71.1%
150Claude Sonnet 4.516k· Epoch AI Benchmarking Hub data71.1%

Showing 101–150 of 302 results.

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence)