in the loop

GPT-6 Sol
#7 of 274 scored.

OpenAISep 22, 2026Closed

Capabilities Index

Range 160.1–166.6

162.7

Context

Takes file, image, text

1.05M

Input

Per 1M tokens

$2.00

Output

Per 1M tokens

$10

Benchmarks

How it scores.

  • GPQA Diamond92.4%

    8th of 186

  • FrontierMath89.8%

    5th of 81

  • ARC-AGI-289.6%

    7th of 79

  • SimpleQA Verified60.7%

    14th of 77

  • Mock AIME100%

    1st of 176

Everything else Epoch has for it

  • DTBench95.5%
  • ARC-AGI95.5%
  • FrontierMath-Tier-4-v2-Private90.0%
  • ProofBench83.0%
  • LMCA69.5%
  • APEX-Agents54.3%
  • EBR-bench53.3%
  • Mystery Game Puzzles51.5%
  • FrontierCode49.3%
  • Furniture Assembly40.5%

On the index

Around it.

  1. #4Claude Sonnet 5.5Anthropic · $2.00 in165.0
  2. #5Claude Fable 5.1Anthropic · $10 in164.7
  3. #6Claude Opus 5Anthropic · $5.00 in162.8
  4. #7GPT-6 SolOpenAI · $2.00 in162.7
  5. #8GPT-5.5 ProOpenAI · $30 in162.1
  6. #9Claude Fable 5Anthropic · $10 in162.1
  7. #10GPT-5.6 SolOpenAI · $2.00 in161.7

Sources