in the loop

Gemini 2.5 Flash (May 2025)
#109 of 274 scored.

GoogleMay 20, 2025Closed

Capabilities Index

Range 139.2–142.9

141.5

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

How it scores.

  • ARC-AGI-22.5%

    61st of 79

  • Humanity's Last Exam6.5%

    26th of 40

  • Mock AIME70.8%

    92nd of 176

Everything else Epoch has for it

  • Fiction.LiveBench77.8%
  • GeoBench76.0%
  • Aider polyglot55.1%
  • WeirdML40.9%
  • ARC-AGI33.3%

On the index

Around it.

  1. #106o1OpenAI · $15 in141.9
  2. #107Gemma 4 26B A4BGoogle · $0.068 in141.8
  3. #108Claude Sonnet 4Anthropic · $3.00 in141.7
  4. #109Gemini 2.5 Flash (May 2025)Google141.5
  5. #110Mistral Medium 3.5Mistral · $1.50 in141.3
  6. #111DeepSeek-R1 (May 2025)DeepSeek141.3
  7. #112Claude 3.7 SonnetAnthropic141.2

Sources