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.

← Integrity Index

ExploitBench

ExploitBench

Rank #18 of 51 · 8 models scored · top score 41

AINTEGRITY 89 / 100

unknown

Strongest on contamination resistance, weakest on discrimination. Caveat: the public test set leaves contamination on the table.

consistent harnesspublic test set

Headroom to ceiling

59 headroomtop 41ceiling 100

Score distribution

median 18GPT-5.5 41112743

State of the art over time

1127434120252026

Integrity breakdown

Weighted composite. Each signal is computed from the source-backed leaderboard below.

Discrimination30%70

Does it still separate models?

Saturation headroom30%95

How far from ceiling / clustered at the top?

Contamination resistance15%100

Public vs held-out; training-leak risk.

Harness comparability15%100

Apples-to-apples, or mixed / vendor-optimized?

Freshness10%100

How old is the benchmark?

Leaderboard

#ModelScoreEvidenceMeasured
1GPT-5.541unverified2026-04-23
2Claude Opus 4.728unverified2026-04-16
3Gemini 3.1 Pro26unverified2026-02-19
4Claude Sonnet 4.624unverified2026-02-17
5Kimi K2.618unverified2026-04-20
6GLM-5.118unverified2026-04-07
7Claude Haiku 4.514unverified2025-10-15
8MiniMax-M2.713unverified2026-03-18

At a glance

Models scored
8
Top model
GPT-5.5
Top score
41
Score spread
28
Headroom
59

Benchmark context

Task type
agentic
Metric
normalized performance in Epoch CSV, rendered as accuracy (%)
Score ceiling
100
Human baseline
unknown
Test-set privacy
unknown
Construction
unknown
Key assumptions
unknown
Limitations
unknown
Contamination history
unknown
Harness notes
Epoch CSV reports model, model version, performance, optimized flag, score source, benchmark release date, and model date.
Authors
unknown
Released
unknown
License
CC BY 4.0
Source
epoch.ai/data/eci_benchmarks.csv

Provenance

8 source records · scores from epoch_benchmarks. Every fact on this page is evidence-linked.

Follow the record

Integrity neighbors

  • The Agent CompanyA · 92
  • GSO-BenchA · 90
  • BalrogA · 89
  • GeoBenchA · 89
  • PostTrainBenchA · 89
  • OTIS Mock AIME 2024-2025A · 88

Top models here

  • GPT-5.541
  • Claude Opus 4.728
  • Gemini 3.1 Pro26
  • Claude Sonnet 4.624
  • Kimi K2.618

Head-to-head

  • GPT-5.5 vs Claude Opus 4.7
  • GPT-5.5 vs Gemini 3.1 Pro
  • Claude Opus 4.7 vs Gemini 3.1 Pro