Singularity Logs

BENCHMARK EXPLORER · math

FrontierMath Tiers 1–3 v2

FrontierMath Tiers 1–3 v2 results published in Epoch AI’s benchmark data.

What it measures

Mean score in the named Epoch AI evaluation; reasoning configurations remain separate.

Interpretation caveats

Scores apply to the reported benchmark and exact model configuration. Other evaluators, dataset revisions and tool settings may differ.

Original benchmark
Family
math
Published versions
1
Metric
Accuracy
Evidence
Source-reported

Published results

Models ordered by FrontierMath Tiers 1–3 v2. Click a row to open its model profile.
#ModalitiesCompare
1OpenAI
InputFileImageTextOutputText
1,050,000tokens
1OpenAI
InputFileImageTextOutputText
1,050,000tokens
2Anthropic
InputTextImageFileOutputText
1,000,000tokens
3OpenAI
InputFileImageTextOutputText
1,050,000tokens
4Anthropic
InputTextImageFileOutputText
1,000,000tokens
5OpenAI
InputFileImageTextOutputText
1,050,000tokens
6OpenAI
InputFileImageTextOutputText
1,050,000tokens
7Anthropic
InputTextImageFileOutputText
1,000,000tokens
8OpenAI
InputFileImageTextOutputText
1,050,000tokens
9Anthropic
InputTextImageFileOutputText
1,000,000tokens
10OpenAI
InputFileImageTextOutputText
1,050,000tokens
10Google DeepMind
InputAudioFileImageTextVideoOutputText
1,048,576tokens
11Anthropic
InputTextImageFileOutputText
1,000,000tokens
12OpenAI
InputFileImageTextOutputText
1,050,000tokens
13OpenAI
InputTextImageFileOutputText
1,050,000tokens
14OpenAI
InputFileImageTextOutputText
1,050,000tokens
14Meta AI
InputTextImageVideoFileAudioOutputText
1,048,576tokens
15Anthropic
InputTextImageFileOutputText
1,000,000tokens
16OpenAI
InputFileImageTextOutputText
1,050,000tokens
17OpenAI
InputTextImageFileOutputText
1,050,000tokens
18Anthropic
InputTextImageFileOutputText
1,000,000tokens
19Alibaba
InputTextOutputText
—Not documented
20Meta AI
InputTextImageVideoFileOutputText
1,048,576tokens
21Meta AI
InputTextImageVideoFileOutputText
1,048,576tokens
22OpenAI
InputImageTextFileOutputText
400,000tokens
23Moonshot
InputTextImageVideoOutputText
1,048,576tokens
24Google DeepMind
InputTextImageVideoFileAudioOutputText
1,048,576tokens
25DeepSeek
InputTextImageOutputText
1,048,576tokens
26Anthropic
InputTextImageFileOutputText
1,000,000tokens
27Meta AI
InputTextImageVideoFileAudioOutputText
1,048,576tokens
28Z.ai (Zhipu AI)
InputTextOutputText
1,048,576tokens
29Google DeepMind
InputTextImageVideoFileAudioOutputText
1,048,576tokens
30OpenAI
InputFileImageTextOutputText
400,000tokens
31DeepSeek
InputTextImageOutputText
1,048,576tokens
32Anthropic
InputTextImageFileOutputText
1,000,000tokens
32xAI
InputTextImageFileOutputText
500,000tokens
33Anthropic
InputTextImageFileOutputText
1,000,000tokens
33Alibaba
InputTextImageVideoOutputText
1,000,000tokens
34DeepSeek
InputTextOutputText
1,048,576tokens
34Alibaba
InputTextOutputText
1,000,000tokens
35Google DeepMind
InputTextImageVideoFileAudioOutputText
1,048,576tokens
36Google DeepMind
InputAudioFileImageTextVideoOutputText
1,048,576tokens
37Z.ai (Zhipu AI)
InputTextOutputText
1,048,576tokens
38Google DeepMind
InputTextImageVideoFileAudioOutputText
1,048,576tokens
39Mistral AI
InputTextImageOutputText
1,048,576tokens
40DeepSeek
InputTextOutputText
1,048,576tokens
41xAI
InputTextImageFileOutputText
500,000tokens
41Moonshot
InputTextImageOutputText
262,144tokens
42Z.ai (Zhipu AI)
InputTextImageVideoOutputText
1,048,576tokens
42OpenAI
InputImageTextFileOutputText
400,000tokens
Source-backed evidence. Different tasks. A clearer picture.
AboutMethodologyBenchmark dashboard
FrontierMath Tiers 1–3 v2 — Singularity Logs