gpt-5.6-luna

StreamingTool callingVisionReasoningJSON modeWeb search

Integration guide

Overview

gpt-5.6-luna is available on MAX API through the OpenAI-compatible API.

Pricing

Official list prices in USD. Token prices are per 1M tokens.

Input ≤ 272,000 tokens

Input: $0.2
Output: $1.2
Cache read: $0.02
Cache write: $0.25

Input > 271,999 tokens

Input: $0.4
Output: $1.8
Cache read: $0.04
Cache write: $0.5

Tier	Input	Output	Cache read	Cache write
Input ≤ 272,000 tokens	$0.2	$1.2	$0.02	$0.25
Input > 271,999 tokens	$0.4	$1.8	$0.04	$0.5

This model uses tiered pricing: the tier is chosen by the total input length of each request.

Cache read applies to prompt tokens served from the prompt cache; cache write applies to tokens written into it.

Example request

curl https://<your-endpoint>/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.6-luna",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

https://<your-endpoint> is a placeholder. Your endpoint is shown in the console after you sign in.