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

DeepSeek · released 2026-10-17

DeepSeek V4.1 Flash 1017

Measured on 115 benchmarks across 3 reasoning efforts.

32.6

#21 of 68 models · best at max

Reasoning efforts in this ranking

max32.6#53 of 150
high28.6#74 of 150
none18.5#113 of 150

Strengths

  • DeepSeek V4.1 Flash 1017
  • 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%28.5%#15 of 68
  2. Terminal-Bench v4-AA10%33.8%#21 of 68
  3. AA-Briefcase v1.115%1570#15 of 68
  4. AutomationBench-AA5%66.6%#14 of 68
  5. Terminal-Bench-Science-AA15%12.9%#18 of 68
  6. HLE-Diamond10%25.3%#25 of 68
  7. CritPt v2 Hard Subset10%25.0%#42 of 68
  8. AA-Omniscience v1.1 Accuracy10%39.6%#25 of 68
  9. AA-Omniscience v1.1 Non-hallucination5%35.9%#53 of 68
  10. GDP.pdf5%17.2%#31 of 68

Bars show where DeepSeek V4.1 Flash 1017 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

cheapest of 50

freeat max

Time per task

#22 quickest of 49

14 minat max

Output speed

#9 fastest of 60

172 tok/sat max

Time to first token

#8 quickest of 60

1.4sat max

Output tokens per task

#46 fewest of 53

139kat 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.

API default effort: high, per Artificial Analysis.

Token price

$0.019 per 1M tokens at API prices

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

245 results
  1. #10high of 24568.5%
  2. #18max of 24566.6%
  3. #98none of 24548.8%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

250 results
  1. #25max of 2501570
  2. #61none of 2501425
  3. #65high of 2501416

Leader: Sonnet 5.5 · 1813

reasoning

CritPt v2 Hard Subset

116 results
  1. #71high of 11627.5%
  2. #76max of 11625.0%
  3. #109none of 1161.0%

Leader: GPT-6 Astra · 72.0%

coding

FrontierCode 2.0-AA

113 results
  1. #34max of 11328.5%
  2. #66high of 11316.8%
  3. #89none of 1136.1%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

243 results
  1. #97max of 24317.2%
  2. #117high of 24314.2%
  3. #188none of 2434.4%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

153 results
  1. #51max of 15325.3%
  2. #66high of 15320.8%
  3. #118none of 1537.1%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

123 results
  1. #59max of 12339.6%
  2. #63high of 12338.9%
  3. #101none of 12321.9%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

123 results
  1. #19none of 12374.7%
  2. #78high of 12338.8%
  3. #83max of 12335.9%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

126 results
  1. #46max of 12612.9%
  2. #52high of 12610.5%
  3. #86none of 1262.4%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

241 results
  1. #46max of 24133.8%
  2. #59high of 24128.8%
  3. #108none of 2419.1%

Leader: Kimi K3.1 · 67.2%

All 115 benchmarks

show 105 more