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.

← Compare

Claude Opus 4.6 vs Claude Sonnet 4.6

Verdict

Claude Opus 4.6 leads 12–1 across 13 shared benchmarks.

Claude Opus 4.6 12 · Claude Sonnet 4.6 1 · higher isn't always better — see caveats.

Per-benchmark head-to-head

Claude Opus 4.6Claude Sonnet 4.6
  • 31.7
    APEX-Agents
    23.7
  • 94
    ARC-AGI
    86.5
  • 69.17
    ARC-AGI-2
    60.42
  • 17
    Chess Puzzles
    13
  • 55.31
    DeepResearch Bench
    54.87
  • 44.12
    GBAEval
    48.76
  • 87.37
    GPQA diamond
    83.16
  • 94.44
    OTIS Mock AIME 2024-2025
    85.78
  • 24.82
    PostTrainBench
    16.42
  • 46.49
    SimpleQA Verified
    29
  • 78.72
    SWE-Bench verified
    75.21
  • 79.8
    Terminal Bench
    53.4
  • 77.95
    WeirdML
    66.07
A higher number is not always a better model. Each score is task performance under a disclosed harness, and some of these benchmarks are saturated or scored under mixed harnesses. Follow any benchmark to its integrity grade before reading a win as decisive.

Follow the record

These models

  • Claude Opus 4.694.44
  • Claude Sonnet 4.686.5