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 2026-06-30

Sonnet 5

Measured on 171 benchmarks across 5 reasoning efforts.

25.0

#39 of 68 models · best at max

Reasoning efforts in this ranking

max25.0#86 of 150
xhigh21.0#103 of 150
high19.1#109 of 150
medium16.6#119 of 150
low14.2#126 of 150

Strengths

  • Sonnet 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%17.8%#34 of 68
  2. Terminal-Bench v4-AA10%14.1%#36 of 68
  3. AA-Briefcase v1.115%1356#36 of 68
  4. AutomationBench-AA5%36.5%#45 of 68
  5. Terminal-Bench-Science-AA15%6.0% est.#29 of 68
  6. HLE-Diamond10%19.1% est.#35 of 68
  7. CritPt v2 Hard Subset10%33.5% est.#30 of 68
  8. AA-Omniscience v1.1 Accuracy10%32.1% est.#39 of 68
  9. AA-Omniscience v1.1 Non-hallucination5%52.4% est.#41 of 68
  10. GDP.pdf5%13.2%#39 of 68

Bars show where Sonnet 5 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

Output speed

#27 fastest of 60

103 tok/sat max

Time to first token

#54 quickest of 60

2.7 minat max

Token price

$0.271 per 1M tokens at API prices$0.0047 on Claude Ultra

97.8% cache read · 0.25% fresh input · 1.73% cache write · 0.27% output

Measured from T3 Code usage in Claude Code, Oct 30 to Nov 5, medium confidence. One app’s users, not everyone’s. $0.294 at the standard 97 / 2.5 / 0.5 mix.

Not measured for Sonnet 5: 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

245 results
  1. #121max of 24536.5%
  2. #129xhigh of 24534.5%
  3. #133high of 24532.1%
  4. #141medium of 24527.9%
  5. #156low of 24519.9%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

250 results
  1. #75max of 2501356
  2. #89xhigh of 2501272
  3. #111high of 2501177
  4. #132medium of 2501058
  5. #155low of 250927

Leader: Sonnet 5.5 · 1813

coding

FrontierCode 2.0-AA

113 results
  1. #61max of 11317.8%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

243 results
  1. #123max of 24313.2%
  2. #130xhigh of 24312.6%
  3. #135medium of 24311.6%
  4. #154low of 2439.4%
  5. #155high of 2439.4%

Leader: Muse Wonder Full · 32.6%

coding

Terminal-Bench v4-AA

241 results
  1. #91max of 24114.1%
  2. #115xhigh of 2417.1%
  3. #122high of 2415.1%
  4. #133low of 2412.5%
  5. #139medium of 2412.0%

Leader: Kimi K3.1 · 67.2%

All 171 benchmarks

show 166 more