- Good as a general-purpose model for text generation and everyday reasoning.
- Good when you want package quota first, with pay-as-you-go enabled by key configuration.
OpenAI active
GPT-5.6 Luna API Pricing and Package Multipliers
Model
gpt-5.6-luna chat.completionsresponses
Efficient GPT-5.6 model for cost-sensitive agent and long-context workloads.
GPT-5.6 Luna API Pricing
Official base: Input 0.6 / Output 3.6 / Cache read 0.06 / Cache write 0.75
| Channel | Billing unit | Quota multiplier | Final discount |
|---|---|---|---|
| Azure | USD / 1M | 6.384x | 5% off |
Cache hits are billed at the cache read price. Cache writes use the cache write price, and the Console usage detail shows the final cost breakdown.
GPT-5.6 Luna available packages and multipliers
| Package | Channel | Billing unit | Quota multiplier | Final discount |
|---|---|---|---|---|
| Super Ultra | Azure | USD / 1M | 63.479x | 10.2% off |
| Ultra | Azure | USD / 1M | 63.479x | 8.2% off |
| SU500 | Azure | USD / 1M | 63.479x | 15.3% off |
| SU750 | Azure | USD / 1M | 63.479x | 16.4% off |
| SU1000 | Azure | USD / 1M | 63.479x | 15.3% off |
| Max | Azure | USD / 1M | 63.479x | 1.11× official |
GPT-5.6 Luna OpenAI-compatible API access
After creating a key in the console, call this model with the unified model name gpt-5.6-luna. When package and pay-as-you-go balance are both available, package quota is used first.
model: "gpt-5.6-luna"
- Not ideal for legacy integrations that only accept OpenAI-compatible APIs.
- Not ideal when you need dedicated SLA terms, private capacity, or a fixed-cost enterprise commitment.
This model is available through the public catalog and unified gateway pricing configuration.
Open this model in the console
The public model page is for discovery and comparison. Actual API access, key creation, and model scopes are managed after sign-in.
Models worth comparing
GPT-5.6 Sol
GPT-5.6 flagship model for frontier reasoning, coding agents, and long-context production workloads.
GPT-5.6 Terra
Balanced GPT-5.6 model for high-quality agent and long-context workloads.
GPT-5.4 Mini
OpenAI lightweight model balancing speed and cost for daily Q&A, batch jobs, and low-latency use cases.