Skip to content
aiqbench

DeepSeek · model passport

DeepSeek V4 Pro High Preview

deepseek-v4-pro-high-preview
  • Open weights
  • MIT licence
  • Released 2026-10-13(first seen on LMArena)
6/27axes covered
Overall · LMArena
1,455#62 of 396
Best percentile
85.6Instruction following
Cohort standing
66.6median 60.6

What the data says

Benchmarked on 6 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 85.6 pct — Instruction following (1,448)

Uneven: 29.5 percentile points between its best and worst axis — 85.6 pct on Instruction following versus 56.1 pct on WebDev Arena.

  • Instruction following85.6 pct1,448Verify →
  • Overall84.6 pct1,455Verify →
  • Hard prompts82.6 pct1,474Verify →
  • Coding77.2 pct1,488Verify →
  • Factuality Arena63.4 pct1,447Verify →
  • WebDev Arena56.1 pct1,464Verify →
  • Vision ArenaNot benchmarked
  • Search ArenaNot benchmarked
  • Document ArenaNot benchmarked
  • Agent ArenaNot benchmarked
  • GPQA DiamondNot benchmarked
  • AIME (math)Not benchmarked
  • FrontierMathNot benchmarked
  • SimpleQA VerifiedNot benchmarked
  • ARC-AGI-2Not benchmarked
  • Humanity's Last ExamNot benchmarked
  • SimpleBenchNot benchmarked
  • LiveBenchNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • IFBench (Artificial Analysis)Not benchmarked
  • HHEM Hallucination (Vectara)Not benchmarked
  • SWE-bench VerifiedNot benchmarked
  • Terminal-BenchNot benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

  • deepseek-v4-pro-high-preview stands at 66.6 (raw mean 74.9) — measured on 6 of 27 axes.

    standingtemplateVerify →

  • deepseek-v4-pro-high-preview stands at 66.6 versus a cohort median of 60.6 across 168 models released within 180 days — above the peer median.

    cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →