Support

Settings

Theme

Explore themes

Mode

Library

Looks borrowed from benchmark sites and products. Themes with a light and a dark version follow your mode.

Bring your own theme

Upload a Priors theme file, or have your agent make one: copy the prompt, tell it the look you want, and upload what it returns. With the Priors skill installed it already knows the format.

Custom themes stay in this browser. They travel inside shared chart links and a page's theme, so others see them as you do.

For the builder, Explore and model pages on this device. Published pages keep the theme their owner picked.

Mode

Benchmarks · math

MathArena Apex

MathArena Apex: 12 final-answer competition problems that earlier frontier models (Grok 4, GPT-5, Gemini 2.5 Pro, GLM 4.5) failed, so it is biased against them. Accuracy over repeated attempts.

41 models · 41 results · data as of 2026-10-29 · matharena.ai ↗

Leaderboard

Chart settings
41 of 41 selected
Opus 4.8 max

Score vs subscription cost per task

MathArena Apex · measured cost per task · subscription-priced where plan data exists (Real API Pricing, full allowance used)

Highlight models

Click a point on the chart to highlight that exact effort; click it again to remove it.

27 more in settings
Chart settings
41 of 41 selected

All results

41 rows
#ModelEffortMathArena ApexSubscription $ / taskAPI $ / task
1Opus 4.8max81.3% ±7.8$0.080$4.59
2GPT-5.5xhigh80.2% ±8.0$0.078$1.42
3GPT-5.4 Proxhigh69.8% ±9.2$8.65$8.65
4Kimi K3max65.6% ±9.5$0.086$1.11
5Gemini 3.1 Prohigh60.9% ±6.9$0.41$0.41
6GPT-5.4xhigh54.2% ±7.1$1.03$1.03
7Opus 4.7xhigh40.6% ±9.8$2.10$2.10
8Opus 4.6high34.5% ±6.8$2.36$2.36
9Gemini 3.5 Flashxhigh32.3% ±9.3$0.30$0.30
10DeepSeek V4 Promax28.1% ±9.0$0.070$0.42
11DeepSeek V4 Flashmax27.1% ±9.3$0.003$0.050
12Gemini 3.6 Flashxhigh26.0% ±8.8$0.19$0.19
13Kimi K2.6max24.0% ±8.5$0.32$0.32
14Gemini 3 Prohigh23.4% ±6.0$0.28$0.28
15GLM 5.2max19.8% ±8.0$0.016$0.18
16Gemini 3 Flashhigh15.6% ±5.1$0.10$0.10
17GPT-5.2high13.5% ±4.8$1.00$1.00
18GLM 5.1max11.5% ±4.5$0.27$0.27
19GLM 5high10.9% ±4.4$0.25$0.25
20DeepSeek V3.2high9.4% ±4.1$0.031$0.031
21Kimi K2.5max8.8% ±4.0$0.18$0.18
22Grok 4 Fastnone5.2% ±3.1$0.011$0.011
23Grok 4.1 Fasthigh5.2% ±3.1$0.013$0.013
24Qwen 3 235B5.2% ±3.1$0.052$0.052
25Qwen 3.5 35B-A3Blow4.2% ±2.8$0.085$0.085
26GPT-5high2.1% ±3.1$3.83$3.83
27Grok 4high2.1% ±2.0$0.52$0.52
28Qwen 3 4Bnone2.1% ±2.0$0.008$0.008
29Qwen 3.5 27Bhigh2.1% ±2.0$0.10$0.10
30Sonnet 4.5high1.6% ±1.7$0.38$0.38