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 2026-09-22

Opus 5.5

Measured on 232 benchmarks across 5 reasoning efforts.

58.2

#2 of 61 models · best at max

Reasoning efforts in this ranking

max58.2#2 of 139
xhigh56.8#5 of 139
high51.5#12 of 139
medium47.9#21 of 139
low34.5#46 of 139

Strengths

  • Opus 5.5
  • 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%46.4%#1 of 61
  2. Terminal-Bench v4-AA10%59.6%#4 of 61
  3. AA-Briefcase v1.115%1801#2 of 61
  4. AutomationBench-AA5%69.5%#7 of 61
  5. Terminal-Bench-Science-AA15%59.0%#3 of 61
  6. HLE-Diamond10%67.9%#2 of 61
  7. CritPt v2 Hard Subset10%61.0%#5 of 61
  8. AA-Omniscience v1.1 Accuracy10%65.5%#2 of 61
  9. AA-Omniscience v1.1 Non-hallucination5%49.7%#38 of 61
  10. GDP.pdf5%26.2%#10 of 61

Bars show where Opus 5.5 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

Cost per task

#17 cheapest of 41

$0.15at max

Time per task

#30 quickest of 40

19 minat max

Output speed

#31 fastest of 52

96 tok/sat max

Time to first token

#52 quickest of 52

10 minat max

Output tokens per task

#41 fewest of 44

173kat max

Token price

$0.349 per 1M tokens at API prices$0.0053 on Claude Max 20x

97.9% cache read · 0.04% fresh input · 1.75% cache write · 0.32% output

Measured from T3 Code usage in Claude Code, Oct 27 to Nov 2, high confidence. One app’s users, not everyone’s. $0.394 at the standard 97 / 2.5 / 0.5 mix.

This ranking’s benchmarks

10 of 10 measured

agentic

AutomationBench-AA

241 results
  1. #8max of 24169.5%
  2. #22xhigh of 24165.0%
  3. #29high of 24163.2%
  4. #36medium of 24161.2%
  5. #82low of 24152.9%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

247 results
  1. #2max of 2471801
  2. #3xhigh of 2471761
  3. #6high of 2471682
  4. #16medium of 2471622
  5. #83low of 2471279

Leader: Sonnet 5.5 · 1814

reasoning

CritPt v2 Hard Subset

100 results
  1. #12xhigh of 10063.0%
  2. #13max of 10061.0%
  3. #19high of 10058.0%
  4. #29medium of 10052.5%
  5. #49low of 10033.5%

Leader: GPT-6 Astra · 71.5%

coding

FrontierCode 2.0-AA

97 results
  1. #1max of 9746.4%
  2. #2xhigh of 9743.8%
  3. #17high of 9735.9%
  4. #24medium of 9733.1%
  5. #60low of 9715.8%

Leader: this model · 46.4%

economics

GDP.pdf

240 results
  1. #14high of 24028.8%
  2. #27xhigh of 24026.6%
  3. #30max of 24026.2%
  4. #35medium of 24025.6%
  5. #36low of 24025.6%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

142 results
  1. #2max of 14267.9%
  2. #4xhigh of 14264.0%
  3. #14high of 14255.6%
  4. #18medium of 14251.4%
  5. #28low of 14241.7%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

111 results
  1. #3xhigh of 11165.5%
  2. #4max of 11165.5%
  3. #5high of 11164.8%
  4. #7medium of 11163.9%
  5. #8low of 11163.3%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

111 results
  1. #62max of 11149.7%
  2. #74high of 11136.1%
  3. #76xhigh of 11135.0%
  4. #78low of 11133.7%
  5. #80medium of 11133.1%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

115 results
  1. #4xhigh of 11561.9%
  2. #7max of 11559.0%
  3. #16high of 11549.0%
  4. #19medium of 11543.3%
  5. #34low of 11524.3%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

238 results
  1. #6max of 23859.6%
  2. #7xhigh of 23859.6%
  3. #13high of 23856.6%
  4. #19medium of 23852.5%
  5. #51low of 23831.3%

Leader: Kimi K3.1 · 67.2%

All 232 benchmarks

show 222 more