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

Gemini 3.1 Pro vs Gemini 3 Pro

Verdict

Gemini 3.1 Pro leads 16–1 across 17 shared benchmarks.

Gemini 3.1 Pro 16 · Gemini 3 Pro 1 · higher isn't always better — see caveats.

Per-benchmark head-to-head

Gemini 3.1 ProGemini 3 Pro
  • 33.5
    APEX-Agents
    18.4
  • 98
    ARC-AGI
    75
  • 77.1
    ARC-AGI-2
    31.11
  • 57
    Balrog
    58.1
  • 55
    Chess Puzzles
    31
  • 20.8
    CL-bench
    15.8
  • 17.71
    CritPt
    6.9
  • 92.13
    GPQA diamond
    90.15
  • 22.55
    GSO-Bench
    18.6
  • 43.74
    HLE
    34.37
  • 95.6
    OTIS Mock AIME 2024-2025
    91.38
  • 21.59
    PostTrainBench
    18.12
  • 75.52
    SimpleBench
    71.68
  • 77.3
    SimpleQA Verified
    72.9
  • 75.62
    SWE-Bench verified
    72.93
  • 80.2
    Terminal Bench
    69.4
  • 72.1
    WeirdML
    69.93
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

  • Gemini 3.1 Pro98
  • Gemini 3 Pro91.38

Gemini 3.1 Pro vs …

  • vs PowerMoE-3b
  • vs Claude Fable 5
  • vs GPT-5.5
  • vs Llama 3.1-405B
  • vs phi-3-small 7.4B