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 3.5 Flash vs Gemini 3 Pro

Verdict

Gemini 3.5 Flash leads 9–2 across 11 shared benchmarks.

Gemini 3.5 Flash 9 · Gemini 3 Pro 2 · higher isn't always better — see caveats.

Per-benchmark head-to-head

Gemini 3.5 FlashGemini 3 Pro
  • 49.6
    APEX-Agents
    18.4
  • 92.5
    ARC-AGI
    75
  • 72.08
    ARC-AGI-2
    31.11
  • 50
    Chess Puzzles
    31
  • 13.14
    CritPt
    6.9
  • 90.4
    GPQA diamond
    90.15
  • 95.55
    OTIS Mock AIME 2024-2025
    91.38
  • 72.04
    SimpleBench
    71.68
  • 68.4
    SimpleQA Verified
    72.9
  • 79.34
    SWE-Bench verified
    72.93
  • 62.64
    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 3.5 Flash95.55
  • Gemini 3 Pro91.38