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

Google · released 2026-09-02

Gemini 3.8 Flash

Measured on 233 benchmarks across 3 reasoning efforts.

30.6

#23 of 61 models · best at high

Reasoning efforts in this ranking

high30.6#57 of 139
medium30.0#61 of 139
low22.4#92 of 139

Strengths

  • Gemini 3.8 Flash
  • 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%18.7%#30 of 61
  2. Terminal-Bench v4-AA10%19.7%#33 of 61
  3. AA-Briefcase v1.115%1203#40 of 61
  4. AutomationBench-AA5%59.9%#22 of 61
  5. Terminal-Bench-Science-AA15%10.0%#21 of 61
  6. HLE-Diamond10%33.1%#16 of 61
  7. CritPt v2 Hard Subset10%43.5%#16 of 61
  8. AA-Omniscience v1.1 Accuracy10%50.5%#11 of 61
  9. AA-Omniscience v1.1 Non-hallucination5%44.9%#42 of 61
  10. GDP.pdf5%21.0%#21 of 61

Bars show where Gemini 3.8 Flash 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

#25 cheapest of 41

$0.31at high

Time per task

#20 quickest of 40

12 minat high

Output speed

#21 fastest of 52

137 tok/sat high

Time to first token

#33 quickest of 52

25sat high

Output tokens per task

#27 fewest of 44

99kat high

Token price

$0.191 per 1M tokens at API prices$0.0077 on Google AI Pro

85.0% cache read · 14.5% fresh input · 0.00% cache write · 0.50% output

Assumed mix: no measured T3 Code usage for this model.

This ranking’s benchmarks

10 of 10 measured

agentic

AutomationBench-AA

241 results
  1. #37medium of 24160.9%
  2. #42high of 24159.9%
  3. #119low of 24136.5%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

247 results
  1. #101high of 2471203
  2. #107medium of 2471183
  3. #128low of 2471062

Leader: Sonnet 5.5 · 1814

reasoning

CritPt v2 Hard Subset

100 results
  1. #38medium of 10044.0%
  2. #39high of 10043.5%
  3. #54low of 10031.5%

Leader: GPT-6 Astra · 71.5%

coding

FrontierCode 2.0-AA

97 results
  1. #51high of 9718.7%
  2. #58medium of 9716.5%
  3. #74low of 978.1%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

240 results
  1. #55medium of 24022.8%
  2. #65high of 24021.0%
  3. #84low of 24018.6%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

142 results
  1. #36high of 14233.1%
  2. #41medium of 14229.1%
  3. #51low of 14224.3%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

111 results
  1. #24high of 11150.5%
  2. #26medium of 11149.6%
  3. #31low of 11148.3%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

111 results
  1. #63medium of 11148.5%
  2. #66high of 11144.9%
  3. #82low of 11131.9%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

115 results
  1. #49medium of 11510.5%
  2. #52high of 11510.0%
  3. #59low of 1158.1%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

238 results
  1. #76high of 23819.7%
  2. #77medium of 23819.7%
  3. #103low of 23810.1%

Leader: Kimi K3.1 · 67.2%

All 233 benchmarks

show 223 more