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 2026-07-30

GPT-5.6 Luna

Measured on 161 benchmarks across 6 reasoning efforts.

25.7

#36 of 61 models · best at max

Reasoning efforts in this ranking

max25.7#81 of 139
xhigh21.9#94 of 139
high20.0#98 of 139
medium13.5#117 of 139
low10.3#126 of 139
none7.9#131 of 139

Strengths

  • GPT-5.6 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%19.6% est.#29 of 61
  2. Terminal-Bench v4-AA10%11.6%#37 of 61
  3. AA-Briefcase v1.115%1339#34 of 61
  4. AutomationBench-AA5%50.2%#37 of 61
  5. Terminal-Bench-Science-AA15%3.3%#35 of 61
  6. HLE-Diamond10%23.9% est.#25 of 61
  7. CritPt v2 Hard Subset10%34.0% est.#23 of 61
  8. AA-Omniscience v1.1 Accuracy10%41.3% est.#21 of 61
  9. AA-Omniscience v1.1 Non-hallucination5%23.6% est.#53 of 61
  10. GDP.pdf5%24.0%#14 of 61

Bars show where GPT-5.6 Luna sits among the 61 ranked models on each benchmark (right is best); the tick is the field median.

Cost and speed

against 61 ranked models

Output speed

#27 fastest of 52

112 tok/sat max

Time to first token

#44 quickest of 52

2.4 minat max

Token price

$0.029 per 1M tokens at API prices$0.0014 on ChatGPT Pro 20x

96.9% cache read · 2.51% fresh input · 0.22% cache write · 0.37% output

Measured from T3 Code usage in Codex, Oct 27 to Nov 2, medium confidence. One app’s users, not everyone’s. $0.030 at the standard 97 / 2.5 / 0.5 mix.

Not measured for GPT-5.6 Luna: cost per task, time per task, output tokens per task (no published API price yet).

This ranking’s benchmarks

5 of 10 measured

agentic

AutomationBench-AA

241 results
  1. #91max of 24150.2%
  2. #108xhigh of 24142.6%
  3. #122high of 24135.6%
  4. #144medium of 24123.1%
  5. #169low of 24111.7%
  6. #182none of 2415.0%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

247 results
  1. #74max of 2471339
  2. #89xhigh of 2471267
  3. #109high of 2471170
  4. #150medium of 247933
  5. #177low of 247741
  6. #186none of 247707

Leader: Sonnet 5.5 · 1814

economics

GDP.pdf

240 results
  1. #46max of 24024.0%
  2. #50xhigh of 24023.8%
  3. #59high of 24022.2%
  4. #110medium of 24015.2%
  5. #115low of 24014.2%
  6. #172none of 2406.0%

Leader: Muse Wonder Full · 32.6%

reasoning

Terminal-Bench-Science-AA

115 results
  1. #79max of 1153.3%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

238 results
  1. #100max of 23811.6%
  2. #122xhigh of 2383.5%
  3. #131high of 2382.5%
  4. #152none of 2381.0%
  5. #162medium of 2380.5%
  6. #178low of 2380.0%

Leader: Kimi K3.1 · 67.2%

All 161 benchmarks

show 156 more