GPT-5.5
A strong general-purpose frontier model with broad evaluation coverage.
INTELLIGENCE55index
CODING—NOT MEASURED
AGENTIC1494Elo
SPEED—NOT MEASURED
FIT / INTERPRETATION
Where the evidence points.
- General reasoning
- Coding
- Tool use
TECHNICAL CONTEXT
Score needs conditions.
- Context
- Not disclosed
- Input / 1M
- $5
- Output / 1M
- $30
- Access
- Proprietary