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 2026-10-07

Haiku 5.5

Measured on 153 benchmarks across 5 reasoning efforts.

36.5

#17 of 68 models · best at max

Reasoning efforts in this ranking

max36.5#42 of 150
xhigh34.6#49 of 150
high29.9#69 of 150
medium24.6#91 of 150
low18.6#111 of 150

Strengths

  • Haiku 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%34.9%#10 of 68
  2. Terminal-Bench v4-AA10%32.8%#24 of 68
  3. AA-Briefcase v1.115%1574#13 of 68
  4. AutomationBench-AA5%35.4%#46 of 68
  5. Terminal-Bench-Science-AA15%20.0%#16 of 68
  6. HLE-Diamond10%29.6%#19 of 68
  7. CritPt v2 Hard Subset10%43.5%#19 of 68
  8. AA-Omniscience v1.1 Accuracy10%34.7%#36 of 68
  9. AA-Omniscience v1.1 Non-hallucination5%65.8%#25 of 68
  10. GDP.pdf5%20.8%#25 of 68

Bars show where Haiku 5.5 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

Cost per task

#9 cheapest of 50

$0.009at max

Time per task

#29 quickest of 49

16 minat max

Output speed

#4 fastest of 60

236 tok/sat max

Time to first token

#55 quickest of 60

3.0 minat max

Output tokens per task

#52 fewest of 53

226kat 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

$0.018 per 1M tokens at API prices$0.00037 on Claude Ultra

95.8% cache read · 0.23% fresh input · 3.22% cache write · 0.75% output

Measured from T3 Code usage in Claude Code, 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. #124xhigh of 24536.0%
  2. #127max of 24535.4%
  3. #131high of 24533.7%
  4. #136medium of 24528.6%
  5. #149low of 24522.9%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

250 results
  1. #23max of 2501574
  2. #36xhigh of 2501523
  3. #56high of 2501441
  4. #72medium of 2501367
  5. #119low of 2501114

Leader: Sonnet 5.5 · 1813

reasoning

CritPt v2 Hard Subset

116 results
  1. #41xhigh of 11647.5%
  2. #44high of 11645.0%
  3. #45max of 11643.5%
  4. #65medium of 11630.5%
  5. #82low of 11621.5%

Leader: GPT-6 Astra · 72.0%

coding

FrontierCode 2.0-AA

113 results
  1. #21max of 11334.9%
  2. #30xhigh of 11329.9%
  3. #48high of 11323.2%
  4. #68medium of 11316.1%
  5. #85low of 1138.0%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

243 results
  1. #71max of 24320.8%
  2. #89xhigh of 24318.2%
  3. #99high of 24317.2%
  4. #112medium of 24315.2%
  5. #139low of 24311.2%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

153 results
  1. #43max of 15329.6%
  2. #45xhigh of 15328.8%
  3. #58high of 15323.7%
  4. #79medium of 15316.8%
  5. #91low of 15314.1%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

123 results
  1. #75max of 12334.7%
  2. #76xhigh of 12333.8%
  3. #81high of 12331.8%
  4. #83medium of 12331.7%
  5. #87low of 12330.4%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

123 results
  1. #34max of 12365.8%
  2. #45high of 12361.1%
  3. #46xhigh of 12360.8%
  4. #49medium of 12359.6%
  5. #50low of 12359.5%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

126 results
  1. #39max of 12620.0%
  2. #40xhigh of 12618.1%
  3. #55high of 12610.0%
  4. #64medium of 1267.1%
  5. #88low of 1261.9%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

241 results
  1. #50max of 24132.8%
  2. #58xhigh of 24129.3%
  3. #73high of 24121.7%
  4. #86medium of 24115.2%
  5. #97low of 24112.6%

Leader: Kimi K3.1 · 67.2%

All 153 benchmarks

show 143 more