priorsBuilder

OpenAI · released 2026-03-05

GPT-5.4 Pro

Measured on 8 benchmarks.

General frontier

Not ranked in this mix — too little measured data for its benchmarks.

Your mix’s benchmarks

1 of 8 measured

math

FrontierMath Tier 4 (v2)

70 results
  1. #20xhigh of 7058.5%

Leader: GPT-6.1 Sol · 100.0%

All 8 benchmarks

show 7 more

creative

MineBench

72 results
  1. #15xhigh of 721829

Leader: GPT-6 Astra Pro · 2281

math

FrontierMath Tiers 1–3 (v2)

115 results
  1. #13xhigh of 11582.5%

Leader: GPT-6.1 Sol · 93.7%

reasoning

ARC-AGI-1

215 results
  1. #35xhigh of 21594.5%

Leader: Fable 5 · 98.5%

reasoning

ARC-AGI-2

216 results
  1. #34xhigh of 21683.3%

Leader: GPT-6 Astra · 95.0%

reasoning

CritPt

529 results
  1. #18xhigh of 52930.0%

Leader: Muse Wonder Full (Max) · 32.9%

reliability

Tool-call error rate

171 results
  1. #1xhigh of 1710.0%

Leader: this model · 0.0%

reliability

Uptime (7d)

181 results
  1. #71xhigh of 18199.8%

Leader: GPT-5 Pro · 100.0%