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 2026-08-13

DeepSeek V4 Pro 0813

Measured on 106 benchmarks across 2 reasoning efforts.

23.0

#42 of 68 models · best at max

Reasoning efforts in this ranking

max23.0#97 of 150
none12.6#130 of 150

Strengths

  • DeepSeek V4 Pro 0813
  • 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%17.1%#37 of 68
  2. Terminal-Bench v4-AA10%14.1%#36 of 68
  3. AA-Briefcase v1.115%1253#43 of 68
  4. AutomationBench-AA5%56.7%#32 of 68
  5. Terminal-Bench-Science-AA15%5.7%#30 of 68
  6. HLE-Diamond10%19.5%#33 of 68
  7. CritPt v2 Hard Subset10%23.5%#44 of 68
  8. AA-Omniscience v1.1 Accuracy10%45.2%#17 of 68
  9. AA-Omniscience v1.1 Non-hallucination5%5.9%#68 of 68
  10. GDP.pdf5%11.4%#44 of 68

Bars show where DeepSeek V4 Pro 0813 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

#42 cheapest of 50

$1.8at max

Time per task

#36 quickest of 49

23 minat max

Output speed

#44 fastest of 60

70 tok/sat max

Time to first token

#19 quickest of 60

2.2sat max

Output tokens per task

#33 fewest of 53

99kat 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

$0.095 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. #63max of 24556.7%
  2. #128none of 24534.9%

Leader: Gemini 4 Carbon · 77.8%

agentic

AA-Briefcase v1.1

250 results
  1. #97max of 2501253
  2. #154none of 250928

Leader: Sonnet 5.5 · 1813

reasoning

CritPt v2 Hard Subset

116 results
  1. #79max of 11623.5%
  2. #93none of 1168.5%

Leader: GPT-6 Astra · 72.0%

coding

FrontierCode 2.0-AA

113 results
  1. #64max of 11317.1%
  2. #83none of 11310.3%

Leader: Opus 5.5 · 46.4%

economics

GDP.pdf

243 results
  1. #136max of 24311.4%
  2. #197none of 2433.6%

Leader: Muse Wonder Full · 32.6%

reasoning

HLE-Diamond

153 results
  1. #69max of 15319.5%
  2. #115none of 1537.2%

Leader: Gemini 4 Carbon · 69.1%

knowledge

AA-Omniscience v1.1 Accuracy

123 results
  1. #44max of 12345.2%
  2. #66none of 12338.6%

Leader: Fable 5.1 · 68.4%

knowledge

AA-Omniscience v1.1 Non-hallucination

123 results
  1. #121none of 1236.1%
  2. #122max of 1235.9%

Leader: Gemini 4 Carbon · 95.8%

reasoning

Terminal-Bench-Science-AA

126 results
  1. #68max of 1265.7%
  2. #103none of 1260.0%

Leader: Gemini 4 Carbon · 64.8%

coding

Terminal-Bench v4-AA

241 results
  1. #90max of 24114.1%
  2. #141none of 2411.5%

Leader: Kimi K3.1 · 67.2%

All 106 benchmarks

show 96 more