GPT-6 Luna
Lowest-cost OpenAI. Simple, high-volume text and image tasks where price matters most.
Which reasoning effort setting?
controls how long GPT-6 Luna thinks before answering. Token prices never change; higher settings spend more and raise task cost.
Start at Max. The top setting is also the best value here. Rule: cheapest setting within 2 points of the model’s best score.
At Max, GPT-6 Luna thinks for about 57.6K tokens per task, then writes about 17.7K more as its answer: $0.15 per task, 29.8× Low, for +18.2 points.
Output rate: $0.50/1M at every effort level. Thinking tokens are billed as output.
| Effort | Index | Per task | Thinking | All output | × Low | Time |
|---|---|---|---|---|---|---|
| None | 9.0 | $0.071 | — | — | — | 2 min |
| Low | 9.7 | $0.0051 | 702 | 3.3K | 1.00× | 0.6 min |
| Medium | 18.5 | $0.042 | 11K | 19.8K | 8.27× | 3.6 min |
| High | 21.4 | $0.073 | 21.6K | 32.8K | 14.29× | 5.9 min |
| Xhigh | 23.5 | $0.10 | 30K | 43.9K | 20.39× | 7.4 min |
| MaxStart here | 27.9 | $0.15 | 57.6K | 75.3K | 29.80× | 12.2 min |
Per task of the Artificial Analysis Intelligence Index: the average over its nine evaluations, counting input, cache, and output tokens at list prices. These are long, agent-style benchmark tasks, much longer than a chat message.
Where GPT-6 Luna sits on the frontier
Intelligence Index against one 100K-input, 10K-output request.