o3
o3 is the recommended small reasoning model in the o-series, offering improved performance, faster responses, and a range of reasoning modes.
Specifications
Context200,000
Max Output100,000
Inputtext, image
Outputtext, json
Performance (7-day Average)
Uptime
TPS
RURT
API Paths
/v1/chat/completions
/v1/responses
Pricing
Input$2.00× 1.1/ MTokens
Output$8.00× 1.1/ MTokens
Cached Input$0.50× 1.1/ MTokens