GPT-6 Luna — API Pricing & Benchmarks
GPT-6 Luna is OpenAI's small reasoning model, released in 2026-09. The API costs $0.10 per 1M input tokens and $0.50 per 1M output tokens ($0.20 blended at a 3:1 ratio), with cached input at $0.01. It supports a 1.1M-token context window. On the independently measured Artificial Analysis Intelligence Index it scores 37.3, ranking #27 of 72 models tracked on this site, with a median output speed of 139 tokens/sec.
Explore provider routes and additional benchmark sources for GPT-6 Luna
| Provider | OpenAI |
| Input price / 1M tokens | $0.10 |
| Output price / 1M tokens | $0.50 |
| Cached input / 1M tokens | $0.01 |
| Blended price (3:1) | $0.20 |
| Context window | 1.1M |
| Max output tokens | 128K |
| Knowledge cutoff | 2026-05 |
| License | proprietary |
| Input modalities | T+I |
| Output speed (tok/s) | 139 |
| Released | 2026-09 |
| Artificial Analysis Intelligence Index | 37.3 |
| Humanity's Last Exam | 38.5% |
Where GPT-6 Luna fits
On a blended 3:1 basis GPT-6 Luna costs $0.20 per 1M tokens, cheaper than 95% of the 84 priced models in this index. It is built for high-volume, latency-sensitive work — classification, extraction, routing and autocomplete — where per-token price matters more than peak intelligence. Cached input is priced at $0.01 (90% below standard input), which rewards prompt structures with long stable prefixes — see our caching guide.
Price history
| Period | Input $/1M | Output $/1M | Cached $/1M |
|---|---|---|---|
| 2026-09-22 → current | $0.10 | $0.50 | $0.01 |
No list-price changes recorded since tracking began. Full dataset: price-history.json (CC BY 4.0).
Alternatives at a similar price
- GPT-4.1 nano (OpenAI) — $0.18/1M blended, AA Index 7.8
- Qwen3.8-Flash (Alibaba Qwen) — $0.23/1M blended, AA Index 39.8
- GLM-5.3-Flash (Zhipu AI (Z.ai)) — $0.24/1M blended, AA Index 41.8
- GPT-4o mini (OpenAI) — $0.26/1M blended, AA Index 6.7
About OpenAI
OpenAI builds the GPT series. The GPT-6 generation (September 2026) ships in three sizes — Astra (flagship, $10/$50), Sol ($2/$10) and Luna ($0.10/$0.50) — that share a 1.05M-token context window and the same tooling, so teams can move workloads between price points without changing integration code. The GPT-5.6 models remain on sale as the previous generation.