---
title: "GPT-6 Luna — API access"
description: "GPT-6 Luna through Codex: the lowest-cost GPT-6 model, 872K maximum context and reasoning from none to max. Official $0.10/$0.50 per 1M input/output tokens; $0.05/$0.25 with the flat 50% discount."
url: "https://apitoken.sale/models/gpt-6-luna"
language: "en"
---
# GPT-6 Luna

GPT-6 Luna is the lightweight, low-cost GPT-6 model for high-volume and latency-sensitive work on the OpenAI-compatible endpoint. Choose reasoning effort from none to max, send images, and use structured outputs or function tools.

- **Model ID:** `gpt-6-luna`
- **Tier:** Fast
- **Context window:** 872K tokens
- **Max output:** 128K tokens
- **Official price (per 1M):** $0.1 input / $0.5 output
- **Your price (per 1M):** input $0.05, output $0.25 (flat 50% off)
- **Base URL:** `https://router.apitoken.sale/v1` · **Endpoints:** `POST /v1/responses`, `POST /v1/chat/completions` (Authorization: Bearer)

## Best for

- Classification, extraction and summarization at scale.
- Latency-sensitive chat, routing and tool calls.
- Cheap pre-processing before an Astra call.

## Notes

- This service uses Luna's Codex subscription limits: 872K maximum context. The larger direct OpenAI API window does not apply here.
- Discovery reports 872K total context, 128K output and 744K conservative input.
- Above 272K input tokens, the whole request uses 2× input/cache rates and 1.5× output rates. Fast (priority) uses 2× the applicable API rates.

## Call it

```bash
curl https://router.apitoken.sale/v1/responses \
  -H "Authorization: Bearer $APITOKEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-6-luna", "input": "Reply with exactly: connected"}'
```

## FAQ

**What does GPT-6 Luna cost?**

Official rates per 1M tokens are $0.10 input, $0.01 cached input, $0.125 cache write and $0.50 output. With the normal 50% B2C discount, these are $0.05, $0.005, $0.0625 and $0.25. OpenKeys uses the official rates.

**How is GPT-6 Luna different from GPT-5.6 Luna?**

GPT-6 Luna is a separate, newer model with its own ID and lower official rates ($0.10/$0.50 against $0.20/$1.20). Both stay available; choose by model ID.

**How do I select Luna?**

Set model to gpt-6-luna on https://router.apitoken.sale/v1. Supported reasoning levels are none, low, medium, high, xhigh and max. Use service_tier priority for Fast.

---
API reference: https://apitoken.sale/md/docs · All models: https://apitoken.sale/md/models