DeepSeek V4 Pro (High): benchmark results

DeepSeek V4 Pro evaluated at the high reasoning-effort setting. Provider: DeepSeek. Released 2026-04-23. Access: Open.

Unified ELO 1667 ± 1, rank #195 of 1761 rated models, from 14 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
OTIS Mock AIME 2024-2595.56Accuracy (%)88.7
LisanBench0.21Mean Path Length / Current Maximum85
ALE-Bench1006.08Performance (Self-Refine x1) (self-reported)72.7
OckBench84Accuracy (%)71.8
FutureEval8.73Unified Forecasting Score69.6
Creative Writing (Lechmazur)0.8Mean Score68.1
Epoch AI - Critpt10Score66.3
WebDev Arena1463.8Arena Score57.3
WeirdML46.53Average Score54.1
Chess Puzzles (Epoch AI)13Accuracy (%)47.5
Surface Evolver Bench Pass Rate25Pass Rate (%)46
SWE-rebench41.43Resolved (%)45.8

Interactive version: theaggregate.ai/model?slug=deepseek-v4-pro-high · How It Works · Data refreshed daily, snapshot 2026-09-05.