GPT-5.6 Luna
GPT-5.6 Luna by OpenAI: $0.20 per 1M input and $1.20 per 1M output tokens, 1.1M context. Call it with one OpenAI-compatible API key, compare it with similar models, start free.
GPT-5.6 Luna is served by OpenAI and called through odnoga on /v1/chat/completions, in the same OpenAI-compatible shape every other model on that endpoint uses. It is a reasoning model, so it spends extra output tokens working through a problem before answering — budget for higher output cost on hard tasks. It accepts images alongside text. It supports tool and function calling. It can be forced to return structured JSON. Its 1.1M-token context window sets how much input you can send in one request. Cached input is billed at $0.02 per 1M tokens, so repeated prefixes cost less.
Specification and price
| Vendor | OpenAI |
|---|---|
| Model ID | gpt-5.6-luna |
| Context window | 1.1M tokens |
| Max output | 128K tokens |
| Served on | /v1/chat/completions |
| Capabilities | Reasoning, Reads images, Tool calling, JSON schema, Streaming, Prompt caching, Batch |
| Per 1M tokens | Vendor cost | Your price on Free+7% |
|---|---|---|
| Input | $0.2 | $0.215 |
| Output | $1.20 | $1.29 |
| Cached input | $0.02 | $0.022 |
Vendor cost is the list price per million tokens as recorded in the odnoga catalog; your price applies your plan margin with the same formula that bills every request — see pricing. Pricing
Call it through odnoga
const res = await fetch("https://api.odnoga.com/v1/chat/completions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.ODNOGA_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
model: "gpt-5.6-luna",
messages: [{ role: "user", content: "Hello" }],
}),
});
One endpoint, one key, one bill — /v1/chat/completions takes the same shape whichever vendor is behind it. Swap the model id and odnoga handles keys, routing, limits and cost accounting.
What GPT-5.6 Luna can do
Everything below is what the model itself supports. Ask for something it does not do and odnoga tells you before the request reaches the provider, instead of returning a worse answer.
Works through a problem step by step before answering, which costs more tokens and buys accuracy on hard tasks.
Send a screenshot, a photo or a scan in the same message as your text.
The model can call functions you define and use what they return.
Answers with valid JSON instead of prose you would have to parse.
Answers in the exact shape you specify, so the response fits your types without a fallback path.
Tokens arrive as they are produced, so your interface can answer immediately.
A prompt you send repeatedly is charged at the cached rate, which is a fraction of the input price.
Work that can wait goes through at a lower price.
Finer control
- Structured outputs · The provider guarantees the schema rather than merely aiming for it.
- Parallel tools · Can call several of your functions in one turn instead of one after another.
How to use GPT-5.6 Luna
Reviewed 2026-09-14
Best for
- Cost-sensitive, high-volume workloads — the tier OpenAI positions exactly there.
- Classification, tagging, routing, moderation pre-checks and short rewrites measured in millions of calls.
- Cheap first-pass work in a two-stage pipeline, with a stronger model handling only what Luna flags.
- Long-context reads on a tight budget: the window is the same 1,050,000 tokens as the rest of the family.
Not the right pick when
- Open-ended reasoning, architecture decisions or long agent runs — use Terra or Sol.
- Customer-visible writing where tone and nuance matter.
- Anything your evaluation shows it fails; cheap is only cheap when the answer is usable.
Practical tips through odnoga
- 01Give it one narrow job per prompt and enforce JSON output — small models are most reliable when the task is closed.
- 02Use it as the first stage and escalate on low confidence; odnoga records the cost of both stages per request so you can see whether the split actually pays.
- 03Batch similar calls behind one managed prompt so a change is one version bump instead of scattered edits.
- 04Watch the per-tenant budget: high volume is exactly where a runaway loop becomes expensive.
What a month costs
1,000 calls a month, 10,000 input tokens and 2,000 output tokens each, at your Free price (vendor cost +7%):
| Input (10M tokens) | $2.15 |
|---|---|
| Output (2M tokens) | $2.58 |
| Your cost per month on Free | $4.73 |
Vendor list cost $4.40 + odnoga margin $0.331 (+7%). Cached input or answer reuse lowers it; odnoga records both numbers per request.
If GPT-5.6 Luna is not the one
Models that take the same kind of input, are priced the same way, and do the closest thing to what this one does. Not a quality ranking — odnoga does not benchmark models.
| Model | Context | Input / 1M | Output / 1M | Capabilities |
|---|---|---|---|---|
| GPT-5.6 Luna | 1.1M | $0.2 | $1.20 | Reasoning, Reads images, Tool calling, JSON schema, Streaming, Prompt caching, Batch |
| GPT-5.4 nano | 400K | $0.2 | $1.25 | Reasoning, Reads images, Tool calling, JSON schema, Streaming, Prompt caching, Batch |
| gpt-6-luna | 1.1M | $0.1 | $0.5 | Reasoning, Reads images, Tool calling, JSON schema, Streaming, Prompt caching, Batch |
| GPT-5 mini | 400K | $0.25 | $2 | Reasoning, Reads images, Tool calling, JSON schema, Streaming, Prompt caching, Batch |
Questions
- What is GPT-5.6 Luna for?
- OpenAI positions Luna for cost-sensitive, high-volume workloads — the nano tier of the GPT-5.6 family, with the same context window and reasoning-effort range as its larger siblings.
- Is Luna accurate enough?
- For closed tasks with a clear output schema it often is. Prove it on your own frozen test cases in the evaluation laboratory rather than assuming; the matrix shows exactly which cases fail.
- How much can it save?
- Compare the worked monthly cost on this page with the same volume on Sol. odnoga records the real per-request cost afterwards, so the saving you model is the one you can audit.
All models · Pricing · Docs · Compare models in the evaluation lab