priorsBuilder

DeepSeek · released 2026-04-24

DeepSeek V4 Pro 0424

Measured on 28 benchmarks across 3 reasoning efforts.

General frontier

38.4

#65 of 136 models · best at max

includes estimates: 24% of the mix is measured for this model

Reasoning efforts in your mix

max38.4#158 of 313
high34.5#187 of 313
none16.6#289 of 313

Your mix’s benchmarks

2 of 8 measured

reasoning

Humanity's Last Exam

587 results
  1. #100max of 58737.5%
  2. #114high of 58735.2%
  3. #339none of 5878.2%

Leader: Opus 5.5 · 61.4%

knowledge

AA-Omniscience

526 results
  1. #191high of 526-10.6
  2. #192max of 526-10.7
  3. #271none of 526-29.9

Leader: Opus 5.5 · 46.4

All 28 benchmarks

show 26 more

agentic

AA-AnalystAgent

44 results
  1. #32max of 4418.8%

Leader: Gemini 3.7 Flash · 60.0%

agentic

AA-Briefcase v1.1

224 results
  1. #131max of 224922

Leader: Sonnet 5.5 · 1823

agentic

APEX-Agents-AA

33 results
  1. #19max of 3324.3%

Leader: Gemini 3.5 Flash · 47.1%

agentic

AutomationBench-AA

219 results
  1. #51max of 21956.3%

Leader: Gemini 4 · 77.5%

agentic

EnterpriseOps-Gym-AA

50 results
  1. #28max of 5040.4%

Leader: Fable 5 · 51.1%

agentic

ITBench-AA

45 results
  1. #22max of 4538.3%

Leader: GPT-5.6 Sol · 56.2%

agentic

τ²-Bench Telecom

393 results
  1. #14max of 39396.2%
  2. #33high of 39394.2%
  3. #60none of 39391.2%

Leader: GLM 5.2 · 99.1%

agentic

τ³-Banking

206 results
  1. #65max of 20630.1%
  2. #78high of 20626.2%

Leader: Qwen 3.8 Max · 51.3%

coding

SciCode

226 results
  1. #98max of 22650.8%

Leader: Opus 5.5 · 66.9%

coding

Terminal-Bench Hard-AA

387 results
  1. #31max of 38746.2%
  2. #50high of 38741.7%
  3. #72none of 38736.4%

Leader: GPT-5.6 Sol · 65.9%

coding

Terminal-Bench v2.1-AA

238 results
  1. #84high of 23864.8%
  2. #87max of 23864.0%

Leader: Fable 5.1 · 91.4%

coding

Terminal-Bench v4-AA

216 results
  1. #70max of 21614.6%

Leader: Sonnet 5.5 · 63.6%

composite

AA Intelligence Index v4.3.2

610 results
  1. #115max of 61030.4
  2. #117high of 61030.1
  3. #214none of 61020.8

Leader: Opus 5.5 · 57.6

cost

AA Intelligence Index Cost per Task

177 results
  1. #43max of 177$0.12

Leader: GPT-6 Luna · $0.00

cost

Cost to Run AA Intelligence Index

177 results
  1. #50max of 177$277

Leader: GPT-6 Luna · $10.63

economics

GDP.pdf

210 results
  1. #101max of 21013.4%

Leader: Muse Wonder Full (Max) · 32.6%

economics

GDPval-AA v2.1

288 results
  1. #123max of 2881156
  2. #124high of 2881150

Leader: Opus 5.5 · 1866

instruction

IFBench

399 results
  1. #20max of 39976.5%
  2. #56high of 39971.3%
  3. #199none of 39945.8%

Leader: Grok 4.3 · 83.3%

knowledge

AA-Omniscience Accuracy

526 results
  1. #102max of 52643.0%
  2. #112high of 52641.4%
  3. #172none of 52630.9%

Leader: Fable 5.1 · 67.2%

knowledge

AA-Omniscience Non-hallucination

526 results
  1. #366none of 52612.2%
  2. #377high of 52611.3%
  3. #485max of 5265.9%

Leader: MiniCPM5-1B (Non-reasoning) · 99.1%

knowledge

GPQA Diamond

539 results
  1. #60high of 53990.5%
  2. #85max of 53988.8%
  3. #277none of 53971.7%

Leader: GPT-6 Astra · 96.3%

long-context

AA-LCR v1.1

523 results
  1. #157max of 52374.7%
  2. #206high of 52370.3%
  3. #295none of 52353.0%

Leader: Kimi K3 · 88.7%

long-context

MLCR-AA

94 results
  1. #28max of 9421.1%

Leader: Sonnet 5.5 · 75.0%

reasoning

CritPt

529 results
  1. #100max of 52912.9%
  2. #114high of 52910.0%
  3. #234none of 5290.9%

Leader: Muse Wonder Full (Max) · 32.9%

speed

Latency (Time to First Token)

351 results
  1. #113high of 3511.6s
  2. #115none of 3511.6s
  3. #124max of 3511.6s

Leader: Gemini 2.5 Flash-Lite · 0.3s

speed

Output Speed (tokens/s)

351 results
  1. #163max of 351100.3
  2. #170none of 35195.5
  3. #176high of 35194.2

Leader: Celeris-1 · 1522.5