GPT-5.6 Luna
openai/gpt-5.6-lunaGPT-5.6 Luna is the smallest, fastest tier of OpenAI's GPT-5.6 reasoning family, with text and image input; OpenAI says it roughly corresponds to the nano tier of earlier GPT-5 families. It is designed for high-volume workloads such as classification, extraction, routing, chat and lightweight agent steps where throughput and latency matter. Pick GPT-5.6 Terra or GPT-5.6 Sol when tasks need more capability, and the pro variant when a hard request needs more reasoning work.
- Input price
- $0.18
$0.20USD / 1M tokens - Output price
- $1.08
$1.20USD / 1M tokens - Cache read
- $0.02
$0.02USD / 1M tokens - Cache write
- $0.23
$0.25USD / 1M tokens - Context window
- 1.1M
- Max output
- 128K
Specifications
- Input
- Text, image
- Output
- Text
About this model
GPT-5.6 Luna is the smallest and fastest tier of OpenAI's GPT-5.6 reasoning family. OpenAI says it matches the nano tier of earlier GPT-5 families. It reads text and images, writes text, and has a 1.05M-token context window at $0.20 per 1M input tokens.
When to use it
- Classification, extraction and routing at high volume.
- Lightweight agent steps that need tools and JSON, not deep reasoning.
- Long-context lookups where the price per token matters.
Call it
The model works on Chat Completions (/v1/chat/completions) and on OpenAI's Responses API (/v1/responses).
from openai import OpenAI
client = OpenAI(base_url="https://api.sandbase.ai/v1", api_key="$SANDBASE_API_KEY")
resp = client.chat.completions.create(
model="openai/gpt-5.6-luna",
messages=[{"role": "user", "content": "Explain what an API gateway does."}],
reasoning_effort="low",
max_tokens=512,
)
print(resp.choices[0].message.content)
What works on SandBase
Checked with real calls through SandBase on October 28, 2026.
| Feature | Works | Notes |
|---|---|---|
| Streaming | Yes | stream: true, with usage in the last chunk when stream_options.include_usage is set. |
| Tool calling | Yes | tools and tool_choice; returns tool_calls. |
| Structured output | Yes | response_format: {"type": "json_schema", …} returned valid JSON. |
| Image input | Yes | image_url parts with an https URL or a data: URI. |
| Reasoning | Adjustable | Set reasoning_effort on Chat Completions or reasoning: {"effort": …} on Responses. In our checks "low" used no reasoning tokens on a short question and "high" did. |
Limits and parameters
- Context window: 1,050,000 tokens.
- Maximum output: 128,000 tokens per request (
max_tokens). - The request schema accepts
temperatureandseed; it has notop_porstop.
Example response
Real response to "In two sentences, explain what an API gateway does." through Chat Completions, trimmed.
{
"id": "e578284d-bf7…",
"object": "chat.completion",
"model": "openai/gpt-5.6-luna",
"choices": [
{
"index": 0,
"finish_reason": "stop",
"message": {
"role": "assistant",
"content": "An API gateway acts as a single entry point for client requests, routing them to the appropriate backend services. It can also handle cross-cutting concerns…"
}
}
],
"usage": {
"completion_tokens": 48,
"completion_tokens_details": {
"reasoning_tokens": 0
},
"prompt_tokens": 17,
"prompt_tokens_details": {
"cache_write_tokens": 0,
"cached_tokens": 0
},
"total_tokens": 65
}
}
Pricing
| Tokens | Price per 1M |
|---|---|
| Input | $0.20 |
| Output | $1.20 |
| Cached input (read) | $0.02 |
| Cache write, 5-minute | $0.25 |
Requests with 272,000 or more prompt tokens are billed at $0.40 per 1M input and $1.80 per 1M output tokens.
Reasoning tokens are part of completion_tokens and are billed at the output price.
Related models
- GPT-5.6 Terra: the middle tier of the same family.
- GPT-5.6 Sol: the most capable tier of the same family.
- GPT-5.6 Luna Pro: Luna with more reasoning work per request.
Independent evaluations
Benchmarks
Scores on standardized evaluations, compared with other models on SandBase.
Metrics sourced from Artificial Analysis · Intelligence Index v5 · Updated 2026-11-13
Try the model
Playground
Write a prompt to try this model. Usage is billed at the prices shown.
Try a prompt
Add an image (optional)
Your response will appear here
Choose an example or write a prompt, then click Run.
Start building
Send your first request
Choose either supported API with unified authentication and usage tracking.
# The quoted heredoc keeps Unicode and shell metacharacters unchanged.
result=$(curl --fail-with-body --silent \
-X POST "https://api.sandbase.ai/v1/responses" \
-H "Authorization: Bearer $SANDBASE_API_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'SANDBASE_JSON'
{
"model": "openai/gpt-5.6-luna",
"input": "Hello",
"max_output_tokens": 2048
}
SANDBASE_JSON
)
printf '%s\n' "$result"Input Schema
8 parameters · 1 required · 7 optional
| Parameter | Type | Required | Description |
|---|---|---|---|
messages | object[] | Required | — |
seed | integer | Optional | — |
tools | array | Optional | — |
stream | boolean | Optional | Default: false |
max_tokens | integer | Optional | Max: 128000 |
temperature | number | Optional | Min: 0 · Max: 2 · Default: 1 |
tool_choice | string | Optional | — |
response_format | object | Optional | — |
Output Schema
| Field | Type | Description |
|---|---|---|
id | string | Unique identifier for the completion |
object | string | Object type, e.g. "chat.completion" |
model | string | Model used for the completion |
choices | array | List of completion choices |
choices[].message.content | string | Generated text content |
choices[].finish_reason | string | Reason the generation stopped |
usage.prompt_tokens | integer | Number of tokens in the prompt |
usage.completion_tokens | integer | Number of tokens in the completion |
usage.total_tokens | integer | Total tokens used |
Questions
FAQ
How do I call OpenAI GPT-5.6 Luna through SandBase?
Create a SandBase API key, then send requests with the model ID "openai/gpt-5.6-luna" to Responses API (/v1/responses) or Chat Completions API (/v1/chat/completions) at https://api.sandbase.ai. The request examples on this page show the exact payload.
How much does OpenAI GPT-5.6 Luna cost?
$0.18 per 1M input tokens and $1.08 per 1M output tokens with the current 10% discount (list price $0.20 / $1.20), billed per request from your SandBase balance.
What is the context window of OpenAI GPT-5.6 Luna?
1.1M tokens of context, with up to 128K output tokens per response.
How does OpenAI GPT-5.6 Luna score on benchmarks?
OpenAI GPT-5.6 Luna scores 23.8 on the Artificial Analysis Intelligence Index (v5), ranking #24 of 31 models on SandBase. Source: Artificial Analysis, measured as GPT-5.6 Luna (Max).
Which features does OpenAI GPT-5.6 Luna support?
OpenAI GPT-5.6 Luna supports vision (image input), reasoning, function calling, structured output. See Specifications above for the full list.
Do I need a separate OpenAI account?
No. One SandBase API key and balance gives you access to OpenAI GPT-5.6 Luna and the other models in the catalog; you do not need to sign up with OpenAI separately.
