in the loop

Claude Opus 5.5
#1 of 274 scored.

AnthropicSep 22, 2026Closed

Capabilities Index

Range 164.1–171.7

167.3

Context

Takes text, image, file

1M

Input

Per 1M tokens

$4.00

Output

Per 1M tokens

$20

Benchmarks

How it scores.

  • GPQA Diamond87.5%

    35th of 186

  • FrontierMath91.2%

    3rd of 81

  • ARC-AGI-293.3%

    3rd of 79

  • SimpleQA Verified72.2%

    4th of 77

  • Mock AIME100%

    1st of 176

Everything else Epoch has for it

  • ProofBench100%
  • ARC-AGI98.5%
  • DTBench98.2%
  • FrontierMath-Tier-4-v2-Private95.0%
  • LMCA80.3%
  • MirrorCode77.4%
  • Furniture Assembly76.2%
  • APEX-Agents73.5%
  • EBR-bench71.4%
  • Mystery Game Puzzles68.0%
  • FrontierSWE62.3%
  • FrontierCode54.6%

On the index

Around it.

  1. #1Claude Opus 5.5Anthropic · $4.00 in167.3
  2. #2GPT-6 AstraOpenAI · $10 in166.4
  3. #3GPT-6.1 SolOpenAI · $2.00 in166.1
  4. #4Claude Sonnet 5.5Anthropic · $2.00 in165.0

Sources