in the loop

GPT-5.2 Pro
#28 of 274 scored.

OpenAIDec 11, 2025Closed

Capabilities Index

Range 153.1–158.1

155.4

Context

Takes image, text, file

400K

Input

Per 1M tokens

$21

Output

Per 1M tokens

$168

Benchmarks

How it scores.

  • FrontierMath74.0%

    20th of 81

  • ARC-AGI-254.2%

    29th of 79

Everything else Epoch has for it

  • ARC-AGI90.5%
  • SimpleBench48.9%
  • FrontierMath-Tier-4-v2-Private46.0%

On the index

Around it.

  1. #25Claude Opus 4.7Anthropic · $5.00 in156.3
  2. #26Claude Sonnet 5Anthropic · $2.00 in156.2
  3. #27GLM-5.3Z.ai · $0.039 in155.6
  4. #28GPT-5.2 ProOpenAI · $21 in155.4
  5. #29DeepSeek V4 Pro 0813DeepSeek · $0.66 in155.3
  6. #30Claude Opus 4.6Anthropic · $5.00 in155.2
  7. #31Qwen3.8 Max (0902)Alibaba · $2.00 in155.1

Sources