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 · reasoning

CritPt v2 Hard Subset

The private 40-question hard subset of CritPt v2: the 70 challenges minus the 30 that seven frontier models solved in all five attempts at launch. Code-execution tool in an offline sandbox; accuracy.

45 models · 93 results · data as of 2026-11-06 · artificialanalysis.ai ↗

Leaderboard

Chart settings
45 of 45 selected
GPT-6 Astra medium

Score vs subscription cost per task

Highlight models

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

Chart settings
45 of 45 selected

Heavy line: best score at each price. Thin lines join one model's effort levels. Cost uses each model's median subscription plan, or API price with no plan data. 27 more models under All models in settings.

More from this benchmark

4 published measures

Breakdown

darker is better within each column · click a column to sort
Model
GPT-6 Astra medium908,872123,055127,589667,438
GPT-6.1 Sol max2,077,656156,223597,5341,480,931
Gemini 4 Carbon high73,733,1251,175,5064,060,42367,661,911
Opus 5.5 max30,507,258398,7643,487,51326,587,314
GPT-6.1 Terra max10,774,559203,6521,331,5539,331,077
GPT-6 Sol max8,677,859155,309955,4007,634,549
Fable 5.1 high8,586,844248,2971,245,2727,100,786
GPT-5.6 Sol max10,515,506198,0321,031,0289,350,938
Sonnet 5.5 max59,677,774494,2125,143,00654,052,155
Gemini 4 Argon high41,985,516609,8622,250,46638,702,215
Muse Wonder (Max)38,235,687374,5282,280,90735,346,049
GPT-5.6 Terra max14,926,552209,6251,322,67113,462,769
Opus 5 max30,304,120516,0642,721,62626,899,477
Kimi K3.1 max21,827,525261,4981,651,69919,794,627
Muse Spark 1.3 max45,639,342638,6442,929,27141,983,634
Haiku 5.5 xhigh25,088,885319,6793,124,86321,799,143
Gemini 3.8 Flash high68,635,173776,2471,883,52865,687,870
Grok 4.7 low29,470,202421,9771,387,07527,422,031
GPT-6 Luna max16,104,278179,0651,772,01714,793,738
Kimi K3 max23,567,341267,7721,495,25721,667,352
Grok 4.6 xhigh31,688,109431,8681,832,63429,463,732
Hy426,952,435293,7402,848,90524,018,959
Inkling 2 Small93,421,1851,206,5314,223,07287,941,331
MiMo V2.6 Pro67,497,157450,0094,201,65962,535,477
GLM 5.3 max98,920,363553,5175,825,70292,412,450

All results

93 rows
#ModelEffortCritPt v2 Hard SubsetSubscription $ / taskAPI $ / taskHarness
1GPT-6 Astramedium72.0%$0.017$0.41Stirrup
2GPT-6 Astrahigh71.5%$0.031$0.71Stirrup
3GPT-6 Astraxhigh71.0%$0.036$0.85Stirrup
4GPT-6 Astramax70.5%$0.049$1.14Stirrup
5GPT-6.1 Solmax69.5%$0.013$0.23Stirrup
6GPT-6.1 Solxhigh69.5%$0.010$0.17Stirrup
7Gemini 4 Carbonhigh69.0%$1.39$3.91Stirrup
8Gemini 4 Carbonmedium69.0%$1.10$3.09Stirrup
9GPT-6.1 Solhigh68.5%$0.008$0.13Stirrup
10Opus 5.5max68.0%$0.039$2.58Stirrup
11GPT-6.1 Terramax68.0%$0.013$0.25Stirrup
12Opus 5.5xhigh67.5%$0.015$0.97Stirrup
13GPT-6.1 Solmedium67.0%$0.004$0.071Stirrup
14GPT-6 Solmax66.5%$0.012$0.38Stirrup
15Gemini 4 Carbonlow65.0%$1.09$3.06Stirrup
16GPT-6 Astralow63.0%$0.010$0.24Stirrup
17GPT-6.1 Sollow62.5%$0.003$0.043Stirrup
18Fable 5.1high62.0%$0.37$2.39Stirrup
19GPT-6.1 Terrahigh61.5%$0.003$0.069Stirrup
20GPT-6.1 Terraxhigh60.5%$0.006$0.12Stirrup
21Fable 5.1xhigh60.0%$0.68$4.33Stirrup
22GPT-5.6 Solmax60.0%Stirrup
23Opus 5.5high59.5%$0.006$0.38Stirrup
24Sonnet 5.5max59.0%$0.029$1.93Stirrup
25Fable 5.1medium58.5%$0.22$1.38Stirrup
26Sonnet 5.5xhigh57.5%$0.007$0.45Stirrup
27GPT-6.1 Terramedium56.5%$0.002$0.040Stirrup
28Opus 5.5medium54.0%$0.004$0.25Stirrup
29Fable 5.1max52.5%$0.71$4.55Stirrup
30Gemini 4 Argonhigh52.5%$1.67$2.35Stirrup