priorsBuilder
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

Anthropic · released 2026-07-24

Opus 5

Measured on 230 benchmarks across 4 reasoning efforts.

44.0

#12 of 67 models · best at max

Reasoning efforts in this ranking

max44.0#27 of 149
high42.2#32 of 149
medium39.5#36 of 149
low34.8#46 of 149

Strengths

  • Opus 5
  • Gemini 4 Carbon (#1)
  • Field median
bestField medianFrontierCode 2.0-…Terminal-Bench v4…AA-Briefcase v1.1AutomationBench-AATerminal-Bench-Sc…HLE-DiamondCritPt v2 Hard Su…AA-Omniscience v1…AA-Omniscience v1…GDP.pdf
  1. FrontierCode 2.0-AA15%31.2%#12 of 67
  2. Terminal-Bench v4-AA10%49.0%#12 of 67
  3. AA-Briefcase v1.115%1655#6 of 67
  4. AutomationBench-AA5%56.6%#33 of 67
  5. Terminal-Bench-Science-AA15%28.6%#12 of 67
  6. HLE-Diamond10%44.8%#11 of 67
  7. CritPt v2 Hard Subset10%50.0%#15 of 67
  8. AA-Omniscience v1.1 Accuracy10%60.9%#5 of 67
  9. AA-Omniscience v1.1 Non-hallucination5%40.6%#46 of 67
  10. GDP.pdf5%21.6%#22 of 67

Bars show where Opus 5 sits among the 67 ranked models on each benchmark (right is best); the tick is the field median.

Cost and speed

against 67 ranked models

Cost per task

#23 cheapest of 50

$0.2at max

Time per task

#40 quickest of 49

30 minat max

Output speed

#45 fastest of 59

66 tok/sat max

Time to first token

#44 quickest of 59

52sat max

Output tokens per task

#39 fewest of 53

118kat max

Time per task: Artificial Analysis · computed, not timed: output tokens per Intelligence Index task divided by output speed, without tool time. Method changed Nov 4, 2026.

Token price

$0.729 per 1M tokens at API prices$0.013 on Claude Ultra

96.8% cache read · 1.19% fresh input · 1.71% cache write · 0.31% output

Measured from T3 Code usage in Claude Code, Oct 30 to Nov 5, high confidence. One app’s users, not everyone’s. $0.735 at the standard 97 / 2.5 / 0.5 mix.

This ranking’s benchmarks

10 of 10 measured

agentic

AutomationBench-AA

245 results
  1. #64max of 24556.6%
  2. #74medium of 24554.3%
  3. #78high of 24553.6%
  4. #80xhigh of 24553.2%
  5. #89low of 24551.8%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

250 results
  1. #10max of 2501655
  2. #15xhigh of 2501629
  3. #28high of 2501557
  4. #59medium of 2501431
  5. #103low of 2501208

Leader: Sonnet 5.5 · 1813

reasoning

CritPt v2 Hard Subset

116 results
  1. #37max of 11650.0%

Leader: GPT-6 Astra · 72.0%

coding

FrontierCode 2.0-AA

113 results
  1. #28max of 11331.2%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

243 results
  1. #64max of 24321.6%
  2. #70xhigh of 24321.0%
  3. #79medium of 24320.0%
  4. #81high of 24319.6%
  5. #100low of 24317.2%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

152 results
  1. #26max of 15244.8%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

122 results
  1. #15max of 12260.9%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

122 results
  1. #72max of 12240.6%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

126 results
  1. #31max of 12628.6%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

241 results
  1. #25max of 24149.0%
  2. #28xhigh of 24146.5%
  3. #29high of 24146.0%
  4. #45medium of 24134.3%
  5. #63low of 24126.3%

Leader: Kimi K3.1 · 67.2%

All 230 benchmarks

show 220 more