priors
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 · agentic

Terminal-Bench 4.0

Resolution rate of agent+model systems on 66 professional computer-work tasks in a terminal (5 trials per task, 8-hour timeout). Rows are harness+model pairs.

23 models · 35 results · data as of 2026-11-05 · tbench.ai ↗

Leaderboard

Chart settings
23 of 23 selected
Opus 5.5 max

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
23 of 23 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. 12 more models under All models in settings.

More from this benchmark

2 published measures
Time per task35 results · lower is better

Highlight models

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

Chart settings
23 of 23 selected

13 more models under All models in settings.

Output tokens per task35 results · lower is better

All results

35 rows
#ModelEffortTerminal-Bench 4.0Subscription $ / taskAPI $ / taskTimeOutput tokensHarness
1Opus 5.5max64.8% ±3.1$0.22$141.3 h269kClaude Code
2Sonnet 5.5max61.8% ±2.9$0.34$222.1 h565kClaude Code
3GPT-6 Astramax58.2% ±2.8$0.42$9.9047 min73kCodex
4GPT-6.1 Solmax58.2% ±3.1$0.11$1.9253 min95kCodex
5Fable 5.1max57.9% ±3.8$2.95$191.1 h191kClaude Code
6Fable 5.1xhigh57.9% ±3.4$2.30$1553 min153kClaude Code
7GPT-6 Astrahigh57.9% ±3.0$0.29$6.8835 min42kCodex
8GPT-6 Astraxhigh57.9% ±2.7$0.30$7.1237 min45kCodex
9Fable 5.1high54.5% ±3.4$1.88$1250 min113kClaude Code
10GPT-6 Astramedium54.2% ±2.7$0.25$5.8031 min33kCodex
11Fable 5.1medium53.9% ±3.4$1.34$8.5842 min79kClaude Code
12Opus 5xhigh53.9% ±3.2$0.32$181.3 h179kClaude Code
13Opus 5max51.8% ±3.4$0.31$181.3 h200kClaude Code
14GPT-6 Astralow50.6% ±2.8$0.20$4.7228 min24kCodex
15Opus 5high50.3% ±3.7$0.24$141.1 h144kClaude Code
16GPT-6 Solmax49.4% ±3.2$0.14$4.4044 min93kCodex
17Opus 5medium44.9% ±3.8$0.17$9.6755 min101kClaude Code
18Fable 5max44.5% ±3.9$2.48$221.2 h178kClaude Code
19Fable 5.1low43.3% ±3.6$1.11$7.1538 min60kClaude Code
20GLM 5.3max41.8% ±3.2$0.86$8.271.6 h208kClaude Code
21Grok 4.7xhigh37.6% ±3.5$1.04$111.6 h70kGrok Build
22GPT-5.6 Solmax37.3% ±3.8$0.21$7.7040 min71kCodex
23GLM 5.3 Flashnone35.8% ±3.5free$0.361.7 h135kClaude Code
24Opus 5low34.9% ±3.9$0.13$7.2549 min72kClaude Code
25Qwen 3.8 Maxmax27.0% ±3.8$1.66$142.8 h253kClaude Code
26Opus 4.8max23.6% ±3.6$0.34$201.4 h269kClaude Code
27GPT-5.6 Terramax21.5% ±3.3$0.15$5.2542 min101kCodex
28Grok 4.6high20.3% ±3.1$1.12$1139 min99kGrok Build
29Gemini 3.8 Flashhigh19.1% ±3.4$0.22$5.5433 min206kmini-SWE-agent
30GPT-5.6 Lunamax17.3% ±2.9$0.049$1.051.1 h186kCodex