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 3.5 Sonnet (October 2024) vs DeepSeek-V3

Verdict

Claude 3.5 Sonnet (October 2024) leads 4–3 across 7 shared benchmarks.

Claude 3.5 Sonnet (October 2024) 4 · DeepSeek-V3 3 · higher isn't always better — see caveats.

Per-benchmark head-to-head

Claude 3.5 Sonnet (October 2024)DeepSeek-V3
  • 51.6
    Aider polyglot
    48.4
  • 3.63
    FrontierMath-2025-02-28-Private
    3.02
  • 40.4
    GPQA diamond
    42.05
  • 56.95
    MATH level 5
    64.85
  • 83.07
    MMLU
    82.93
  • 8.38
    OTIS Mock AIME 2024-2025
    15.75
  • 29.68
    SimpleBench
    2.68
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 3.5 Sonnet (October 2024)83.07
  • DeepSeek-V393.73

DeepSeek-V3 vs …

  • vs PowerMoE-3b
  • vs Claude Fable 5
  • vs GPT-5.5
  • vs Gemini 3.1 Pro
  • vs Llama 3.1-405B