NexRelayNexRelay
LoginRegister

Model pricing

Model price comparison

Compare model input, output, and cached token costs to choose a safer routing policy for AI API workloads.

Create account to testView pricing
ModelInput / 1MCached / 1MOutput / 1MBest fit
gpt-5.2$0.60$0.08$2.40Routine requests
gpt-5.3$1.20$0.15$5.00Balanced workloads
gpt-5.4$2.50$0.30$10.00Harder tasks
gpt-5.5$5.00$0.60$20.00Premium short runs
Use lower-cost models for routine work, cap output tokens for coding agents, and reserve premium models for short high-value requests.