GPT-5.6 launches with three tiered models

Summary

Cost-per-inference drops significantly across tiers, directly impacting feasibility of production inference workloads. Predictable cache behavior with explicit breakpoints reduces latency variance in cached query patterns.

Why it matters

Cost-per-inference drops significantly across tiers, directly impacting feasibility of production inference workloads. Predictable cache behavior with explicit breakpoints reduces latency variance in cached query patterns.

Sources

1. 1.Terra has competitive performance to GPT‑5.5 while being 2x cheaper and Luna brings strong capability at our lowest cost
2. 2.GPT‑5.6 is priced per 1M tokens across three model sizes: Sol is $5 input / $30 output; Terra is $2.50 input / $15 output; and Luna is $1 input / $6 output
3. 3.GPT‑5.6 also introduces more predictable prompt caching, including support for explicit cache breakpoints and a 30-minute minimum cache life
4. 4.cache writes are billed at 1.25x the model's uncached input rate, while cache reads continue to receive the 90% cached-input discount