Skip to content

Benchmark

U-MATH differential_calc accuracy

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

Models measured
30
Scores recorded
90
Sources
1
IQ weight
2.13

Score distribution

35.2%
Mean
30.7%
Median
20%
Std dev
11.4%
10th percentile
65.5%
90th percentile
4.3%
Min
4.3%88%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.13 for chat models, measured over 30 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1o3 Minidifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH88%
2o1differential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH86%
3DeepSeek R1differential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH85.3%
4Gemini 2.0 Flash Experimentaldifferential_calc, text, gpt-4o-2024-08-06, thinking· U-MATH and mu-MATH80.7%
5o1differential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH76.8%
6O1 Minidifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH75.3%
7Gemini 2.0 Flash Experimentaldifferential_calc, all, gpt-4o-2024-08-06, thinking· U-MATH and mu-MATH70.5%
8QwQ 32B Previewdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH70%
9O1 Minidifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH68.2%
10DeepSeek R1differential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH65.5%
11o3 Minidifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH65.5%
12o1differential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH57.1%
13QwQ 32B Previewdifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH55.5%
14QVQ 72B Previewdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH54%
15O1 Minidifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH52.9%
16Gemini 1.5 Prodifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH50.7%
17QVQ 72B Previewdifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH50%
18Gemini 1.5 Prodifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH49.6%
19DeepSeek V3differential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH49.3%
20Gemini 2.0 Flash Experimentaldifferential_calc, visual, gpt-4o-2024-08-06, thinking· U-MATH and mu-MATH48.6%
21Gemini 1.5 Flashdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH47.3%
22Gemini 1.5 Flashdifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH47.3%
23Gemini 1.5 Flashdifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH47.1%
24Gemini 1.5 Prodifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH47.1%
25Qwen2.5 Math 72B Instructdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH46%
26Qwen2-VL (72B) Instructdifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH44.3%
27DeepSeek V3differential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH43.2%
28Phi 4differential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH42.7%
29Qwen2.5 32B Instructdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH42.7%
30Phi 4differential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH42.3%
31Claude 3.5 Sonnetdifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH41.4%
32Phi 4differential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH41.4%
33QVQ 72B Previewdifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH41.4%
34GPT 4odifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH37.1%
35Qwen2 VL 7b Instructdifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH37.1%
36Qwen2.5 72B Instructdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH36.7%
37Qwen2.5 Math 72B Instructdifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH35.5%
38Llama 3.3 70B Instructdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH35.3%
39Qwen2-VL (72B) Instructdifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH34.1%
40GPT-4o minidifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH32.9%
41Qwen2.5 32B Instructdifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH32.7%
42GPT 4odifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH32.3%
43Mistral Large Instruct 2411differential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH32%
44Qwen2.5 Math 7B Instructdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH32%
45Qwen2.5 7B Instructdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH30.7%
46DeepSeek V3differential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH30%
47GPT 4odifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH30%
48Pixtral 12Bdifferential_calc, visual, gpt-4o-2024-08-06· U-MATH and mu-MATH30%
49Qwen2.5 72B Instructdifferential_calc, all, gpt-4o-2024-08-06· U-MATH and mu-MATH30%
50Nvidia Llama 3.1 Nemotron 70B Instructdifferential_calc, text, gpt-4o-2024-08-06· U-MATH and mu-MATH29.3%

Showing 1–50 of 90 results.

Data credits: Toloka, U-MATH and mu-MATH public leaderboard. MIT Space; generated solution outputs retain underlying model terms and are not ingested. (licence)