tensor.news
Integrity IndexModelsCompareLabsPapersMethodologyAbout
tensor.news

© 2026

Evidence

  • Integrity Index
  • Model Claim Ledger
  • Compare

Entities

  • Models
  • Benchmarks
  • Labs
  • Papers

Method

  • Methodology
  • About
  • Privacy

Machine

  • Feed (Atom)
  • llms.txt
  • llms-full.txt
  • Sitemap

tensor.news grades benchmarks and tags model claims by evidence status. Every fact carries a source record and provenance. Scores are framed as task performance under a disclosed harness, never as deployed capability.

← Models

GPT-5.4 Pro

OpenAI·released 2026-03-05

independently reproduced

Scored on 11 benchmarks, leading 1. 45% of its scores are independently reproduced — the rest are self-reported or unverified.

11
Benchmarks scored
1
SOTA leads
94.5
Best score
45%
Reproduced
5 of 11

Leads on: GPQA diamond

Benchmark fingerprint

Every score, ranked, colored by evidence status.

  • ARC-AGI
    94.5
  • GPQA diamond
    92.8
  • ARC-AGI-2
    83.33
  • FrontierMath-Tiers-1-3-v2-Private
    82.46
  • SimpleBench
    68.92
  • Chess Puzzles
    58.6
  • FrontierMath-Tier-4-v2-Private
    58.54
  • WeirdML
    57.44
  • SimpleQA Verified
    47.8
  • HLE
    41.51
  • CritPt
    30

All benchmark scores

BenchmarkScoreEvidence statusMeasured
ARC-AGI94.5accuracy (%)unverified· optimized2026-03-05
GPQA diamond92.8accuracy (%)reproduced· optimized2026-03-05
ARC-AGI-283.33accuracy (%)unverified2026-03-05
FrontierMath-Tiers-1-3-v2-Private82.46accuracy (%)reproduced2026-03-05
SimpleBench68.92accuracy (%)unverified2026-03-05
Chess Puzzles58.6accuracy (%)reproduced2026-03-05
FrontierMath-Tier-4-v2-Private58.54accuracy (%)reproduced2026-03-05
WeirdML57.44accuracy (%)unverified2026-03-05
SimpleQA Verified47.8accuracy (%)reproduced2026-03-05
HLE41.51accuracy (%)unverified2026-03-05
CritPt30accuracy (%)unverified2026-03-05
11 source records · epoch_benchmarks. Every score is evidence-linked.

Follow the record

More from OpenAI

  • GPT-5.5100
  • GPT-5.5 Pro100
  • GPT-598.13
  • GPT-5 mini97.85
  • o4-mini97.83
  • o397.77

Benchmarks it leads

  • GPQA diamond