in the loop

Gemini 2.5 Pro (Mar 2025)
#91 of 274 scored.

GoogleMar 31, 2025Closed

Capabilities Index

Range 141.9–147.0

144.2

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

How it scores.

  • GPQA Diamond78.5%

    68th of 186

  • Humanity's Last Exam14.0%

    20th of 40

Everything else Epoch has for it

  • MATH level 595.6%
  • GeoBench81.0%
  • Lech Mazur Writing80.5%
  • Aider polyglot72.9%
  • Fiction.LiveBench66.7%
  • CadEval64.0%
  • Balrog43.3%
  • SimpleBench41.9%
  • ARC-AGI33.0%
  • VPCT22.0%

On the index

Around it.

  1. #88Qwen3.7 FlashAlibaba · $0.03 in144.6
  2. #89Gemini 3.1 Flash-LiteGoogle · $0.25 in144.5
  3. #90Grok 4 FastxAI144.2
  4. #91Gemini 2.5 Pro (Mar 2025)Google144.2
  5. #92Claude Opus 4.1Anthropic · $15 in144.1
  6. #93Qwen 3.5 Flash (hosted 35B-A3B)Alibaba144.0
  7. #94Qwen 3.6 35B-A3BAlibaba · $0.15 in143.9

Sources