GPT-5.6 Luna API cost and pricing
Budget short summaries with this exact model's live input, output and cache prices. The three scenarios below separate usage assumptions from current rates.
gpt-5.6-lunaLive token and cache rates
- Fresh input / 1M tokens
- $0.20
- Output / 1M tokens
- $1.20
- Cached input / 1M tokens
- $0.02 per 1M tokens
- Cache write / 1M tokens
- No extra fee
- Cached output
- Not offered
A cache hit must be reported by the provider. Without a separate cached-input rate, normal input pricing applies. Cache creation uses a listed write rate when applicable; generated output never receives a cached-output discount.
Three monthly workload estimates
Illustrative scenarios, not measured traffic or promised bills. Input includes system instructions, history and tool results. All output, including chargeable reasoning, must fit the assumed output count. These totals assume no cache hits or writes.
| Workload | Input / request | Output / request | Requests / month | Monthly estimate |
|---|---|---|---|---|
| Short question answering | 1,500 | 300 | 10,000 | $7.29 |
| short summaries | 3,000 | 250 | 3,000 | $2.81 |
| Long-context review | 40,000 | 1,500 | 500 | $4.95 |
Formula: round each request's (input tokens × input rate + output tokens × output rate) / 1,000,000 up to the next INR paise, then multiply by request count. Billing stays in INR paise; display currency does not change charges. No GST surcharge is added.
Standard, thinking and Fast variants
Each enabled variant has its own ID. The same per-token rate can still produce different bills when reasoning or answer lengths differ. Fast-serving rates are read from their own rows, never inferred from the base model.
- GPT-5.6 Luna$0.20 in / $1.20 out per 1M tokens
- GPT-5.6 Luna Deep$0.20 in / $1.20 out per 1M tokens
- GPT-5.6 Luna Fast$0.20 in / $1.20 out per 1M tokens
- GPT-5.6 Luna Max$0.20 in / $1.20 out per 1M tokens
What the estimate leaves out
Extra agent steps, growing conversation history, additional billable requests, external tools, multimodal inputs and provider-specific conditions can change totals. Check recorded usage and current prices after testing a representative workload. Use the calculator for explicit cache counts rather than assuming every repeated prompt qualifies.
Model-specific checks
Summary workloads are input-heavy. Compare total source-text tokens as well as output length when budgeting.
- Context capacity
- 1,050,000 tokens
- Maximum output
- 128,000 tokens
- Current usage price
- $0.20 in / $1.20 out per 1M tokens
Limits are provider-published where a source exists, not independently tested TokenRex request limits. Exact-version benchmarks remain on the model page; scores from another variant are not copied here.
Sources and verification
TokenRex's live catalog determines availability and charges. Provider documents describe upstream behavior; they do not prove that every native feature is exposed here. These examples are not a claim of a completed live API test.
- TokenRex enabled catalog and prices
- TokenRex endpoint and billing documentation
- Official OpenAI Python SDK: custom client configuration
- Official OpenAI Node SDK: custom client configuration
- Provider model specifications · reviewed 2026-10-06
- Cognition · FrontierCode 1.1 original leaderboard · reviewed 2026-10-07