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

OpenAI · released 2026-09-22

GPT-6 Luna

Measured on 181 benchmarks across 6 reasoning efforts.

27.9

#33 of 68 models · best at max

Reasoning efforts in this ranking

max27.9#78 of 150
xhigh23.4#94 of 150
high21.4#101 of 150
medium18.4#114 of 150
low9.7#138 of 150
none9.4#140 of 150

Strengths

  • GPT-6 Luna
  • 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%25.2%#21 of 68
  2. Terminal-Bench v4-AA10%12.6%#39 of 68
  3. AA-Briefcase v1.115%1337#39 of 68
  4. AutomationBench-AA5%53.2%#36 of 68
  5. Terminal-Bench-Science-AA15%8.6%#26 of 68
  6. HLE-Diamond10%22.2%#29 of 68
  7. CritPt v2 Hard Subset10%42.5%#20 of 68
  8. AA-Omniscience v1.1 Accuracy10%39.6%#24 of 68
  9. AA-Omniscience v1.1 Non-hallucination5%20.7%#58 of 68
  10. GDP.pdf5%22.8%#18 of 68

Bars show where GPT-6 Luna sits among the 68 ranked models on each benchmark (right is best); the tick is the field median.

Cost and speed

against 68 ranked models

Cost per task

#10 cheapest of 50

$0.01at max

Time per task

#18 quickest of 49

12 minat max

Output speed

#29 fastest of 60

101 tok/sat max

Time to first token

#53 quickest of 60

1.7 minat max

Output tokens per task

#24 fewest of 53

75kat 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

$0.015 per 1M tokens at API prices$0.0011 on ChatGPT Pro 10x

96.6% cache read · 2.78% fresh input · 0.10% cache write · 0.50% output

Measured from T3 Code usage in Codex, Oct 30 to Nov 5, high confidence. One app’s users, not everyone’s. $0.015 at the standard 97 / 2.5 / 0.5 mix.

This ranking’s benchmarks

10 of 10 measured

agentic

AutomationBench-AA

245 results
  1. #81max of 24553.2%
  2. #101high of 24547.8%
  3. #103xhigh of 24547.8%
  4. #116medium of 24540.5%
  5. #171low of 24512.0%
  6. #174none of 2459.1%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

250 results
  1. #78max of 2501337
  2. #100xhigh of 2501235
  3. #108high of 2501187
  4. #123medium of 2501096
  5. #161none of 250903
  6. #181low of 250740

Leader: Sonnet 5.5 · 1813

reasoning

CritPt v2 Hard Subset

116 results
  1. #47max of 11642.5%
  2. #53xhigh of 11637.0%
  3. #64high of 11631.0%
  4. #72medium of 11627.5%
  5. #92low of 1169.5%
  6. #98none of 1165.0%

Leader: GPT-6 Astra · 72.0%

coding

FrontierCode 2.0-AA

113 results
  1. #43max of 11325.2%
  2. #59high of 11318.6%
  3. #60xhigh of 11318.4%
  4. #74medium of 11312.8%
  5. #96low of 1132.2%
  6. #98none of 1131.6%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

243 results
  1. #58max of 24322.8%
  2. #86xhigh of 24318.6%
  3. #90high of 24318.2%
  4. #119medium of 24313.8%
  5. #157low of 2439.0%
  6. #176none of 2435.8%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

153 results
  1. #61max of 15322.2%
  2. #71xhigh of 15318.9%
  3. #83high of 15315.6%
  4. #87medium of 15315.1%
  5. #106low of 15310.2%
  6. #123none of 1536.3%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

123 results
  1. #58max of 12339.6%
  2. #62high of 12339.0%
  3. #64medium of 12338.7%
  4. #65xhigh of 12338.6%
  5. #73low of 12336.1%
  6. #88none of 12330.1%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

123 results
  1. #97max of 12320.7%
  2. #98none of 12320.0%
  3. #103xhigh of 12316.0%
  4. #105low of 12315.2%
  5. #106high of 12315.1%
  6. #107medium of 12314.7%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

126 results
  1. #59max of 1268.6%
  2. #70xhigh of 1265.2%
  3. #84high of 1262.9%
  4. #89medium of 1261.4%
  5. #93low of 1261.0%
  6. #94none of 1261.0%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

241 results
  1. #96max of 24112.6%
  2. #112xhigh of 2418.1%
  3. #123high of 2414.5%
  4. #131medium of 2412.5%
  5. #142none of 2411.5%
  6. #174low of 2410.0%

Leader: Kimi K3.1 · 67.2%

All 181 benchmarks

show 171 more