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

Update notices

A note when a benchmark one of your indexes uses gets a new version, with a one-click upgrade.

Benchmarks · Speed

T3 Code minutes per turn by effort

How long a turn takes in real coding work, by model and reasoning effort: average time from sending a turn in T3 Code to its completion, tool and wait time included, over the last 30 days. Each figure says how many users it rests on; ones from few users are noisy. From T3 Code's own telemetry, shared by Theo.

46 models · 179 results · updated 2026-11-16 · t3.codes ↗

Leaderboard

GPT-5.5

More from this benchmark

6 published measures

Breakdown

darker is better within each column · click a column to sort
Model
GPT-5.562.00.090.28
Opus 5 low90.08.8120.0
Muse Spark 1.3 minimal610.541.5
GPT-5.6 Luna low322.05.4964.7
Opus 4.5 low15.019.092.0
Haiku 5.5 low5,5658.6953.0
Muse Wonder minimal13.024.3150.3
GLM 5.3 Flash low733.6125.5
Sonnet 4.6 low76.018.7139.2
GPT-6.1 Terra low1,66221.6119.2
Haiku 4.51,5805.8792.6
Grok 4.5 low41.021.7139.3
GPT-5.6 Terra low147.020.7171.6
Opus 4.6 low51.018.6210.5
GPT-6 Luna low4,44216.5145.2
Sonnet 5.5 low5,63217.0168.2
Sonnet 5 low87.020.1185.9
Claude Mythos 5.1 medium11.033.2281.7
Opus 4.7 medium75.036.8246.9
Opus 4.8 low35.069.5230.7
GPT-5.6 Sol low314.026.1215.2
Opus 5.5 low10,26222.1197.9
Fable 5 low19.041.8239.6
Grok 4.6 low75.049.6300.3
GPT-6 Sol low310.028.9208.8

All results

179 rows
#ModelEffortScoreTimeOutput tokens
1GPT-5.50.16.5s5.1k
2Opus 5low0.320s720
3Muse Spark 1.3minimal0.321s567
4GPT-5.6 Lunalow0.743s769
5Opus 4.5low0.846s1.7k
6Haiku 5.5low0.847s1.5k
7Muse Wonderminimal0.954s1.4k
8Opus 4.5medium1.059s2.2k
9GLM 5.3 Flashlow1.167s1.0k
10Sonnet 4.6low1.168s2.8k
11GPT-6.1 Terralow1.272s1.3k
12Haiku 4.51.375s1.5k
13Grok 4.5low1.376s218
14GPT-5.6 Lunamedium1.487s2.6k
15GPT-5.6 Terralow1.61.6 min1.2k
16Opus 4.6low1.61.6 min1.8k
17GPT-6 Lunalow1.71.7 min2.7k
18Haiku 5.5medium1.71.7 min7.0k
19Sonnet 5.5low1.81.8 min2.7k
20Sonnet 5low1.81.8 min3.2k
21Claude Mythos 5.1medium1.81.8 min6.2k
22Muse Spark 1.3low1.81.8 min2.2k
23GPT-6.1 Terramedium1.91.9 min2.3k
24Opus 4.7medium1.91.9 min4.5k
25Opus 4.8low1.91.9 min5.0k
26GLM 5.3 Flash2.02.0 min4.5k
27GPT-5.6 Sollow2.02.0 min2.9k
28Opus 5.5low2.02.0 min3.1k
29Fable 5low2.02.0 min3.9k
30Grok 4.6low2.12.2 min3.6k