← Model benchmark
OpenAI · GPT

GPT-5.5

A frontier model for agentic coding, computer use, scientific work, and professional knowledge tasks with broad product and API deployment.

DOOM SCORE80.9out of 100
CURRENT ASSESSMENT · REVISION 1

Why this model scores 80.9

High agentic competence and broad access increased practical reach, while documented preparedness testing and safeguards reduced but did not remove dangerous-domain risk.

Capability88
Autonomy86
Deployment82
Misuse potential76
Control difficulty68
AUDIT TRAIL

Model score history

  1. R1
    Doom Score 80.9

    Initial source-backed model assessment

    11 Aug 2026