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

Llama 3.1-405B vs Qwen2.5-72B

Verdict

Llama 3.1-405B leads 10–2 across 12 shared benchmarks.

Llama 3.1-405B 10 · Qwen2.5-72B 2 · higher isn't always better — see caveats.

Per-benchmark head-to-head

Llama 3.1-405BQwen2.5-72B
  • 93.73
    ARC AI2
    92.67
  • 77.2
    BBH
    73.07
  • 34.55
    GPQA diamond
    32.2
  • 85.6
    HellaSwag
    79.73
  • 49.77
    MATH level 5
    63.17
  • 79.33
    MMLU
    80.4
  • 9.63
    OTIS Mock AIME 2024-2025
    7.96
  • 71.8
    PIQA
    65.2
  • 7.4
    The Agent Company
    5.7
  • 82.7
    TriviaQA
    71.9
  • 21.38
    WeirdML
    15.97
  • 78.4
    Winogrande
    64.6
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

  • Llama 3.1-405B93.73
  • Qwen2.5-72B92.67

Llama 3.1-405B vs …

  • vs PowerMoE-3b
  • vs Claude Fable 5
  • vs GPT-5.5
  • vs Gemini 3.1 Pro
  • vs phi-3-small 7.4B