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

Alibaba · released 2026-08-14

Qwen 3.8 27B

Measured on 136 benchmarks across 4 reasoning efforts.

20.0

#40 of 61 models · best at xhigh

Reasoning efforts in this ranking

xhigh20.0#99 of 139
medium17.4#111 of 139
low15.6#112 of 139
none10.1#127 of 139

Strengths

  • Qwen 3.8 27B
  • 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%14.9%#36 of 61
  2. Terminal-Bench v4-AA10%5.6%#40 of 61
  3. AA-Briefcase v1.115%1393#32 of 61
  4. AutomationBench-AA5%48.2%#40 of 61
  5. Terminal-Bench-Science-AA15%0.5%#41 of 61
  6. HLE-Diamond10%15.3%#38 of 61
  7. CritPt v2 Hard Subset10%4.0%#52 of 61
  8. AA-Omniscience v1.1 Accuracy10%17.3%#55 of 61
  9. AA-Omniscience v1.1 Non-hallucination5%71.4%#15 of 61
  10. GDP.pdf5%16.4%#30 of 61

Bars show where Qwen 3.8 27B 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

#22 cheapest of 41

$0.26at xhigh

Time per task

#37 quickest of 40

32 minat xhigh

Output speed

#48 fastest of 52

41 tok/sat xhigh

Time to first token

#30 quickest of 52

4.0sat xhigh

Output tokens per task

#29 fewest of 44

110kat xhigh

Token price

$0.124 per 1M tokens at API prices$0.0091 on Command Code GOAT

97.0% cache read · 2.50% 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. #98xhigh of 24148.2%
  2. #114medium of 24140.3%
  3. #131low of 24130.1%
  4. #167none of 24112.4%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

247 results
  1. #64xhigh of 2471393
  2. #67medium of 2471373
  3. #87low of 2471270
  4. #159none of 247901

Leader: Sonnet 5.5 · 1814

reasoning

CritPt v2 Hard Subset

100 results
  1. #77medium of 10016.5%
  2. #86low of 1005.5%
  3. #88xhigh of 1004.0%
  4. #92none of 1002.0%

Leader: GPT-6 Astra · 71.5%

coding

FrontierCode 2.0-AA

97 results
  1. #62xhigh of 9714.9%
  2. #67low of 9712.4%
  3. #70medium of 9711.5%
  4. #85none of 973.7%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

240 results
  1. #102xhigh of 24016.4%
  2. #122medium of 24012.8%
  3. #145low of 24010.4%
  4. #179none of 2405.2%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

142 results
  1. #77xhigh of 14215.3%
  2. #112low of 1426.8%
  3. #113medium of 1426.7%
  4. #135none of 1424.2%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

111 results
  1. #99low of 11118.3%
  2. #101medium of 11117.9%
  3. #103xhigh of 11117.3%
  4. #111none of 1119.1%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

111 results
  1. #6none of 11182.0%
  2. #22xhigh of 11171.4%
  3. #57low of 11151.3%
  4. #77medium of 11134.6%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

115 results
  1. #87medium of 1151.0%
  2. #91xhigh of 1150.5%
  3. #94low of 1150.5%
  4. #102none of 1150.0%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

238 results
  1. #115xhigh of 2385.6%
  2. #117medium of 2385.1%
  3. #129low of 2382.5%
  4. #170none of 2380.0%

Leader: Kimi K3.1 · 67.2%

All 136 benchmarks

show 126 more