Pyyan / Compare / GPT-6 Luna vs GPT-6 Astra
GPT-6 Luna vs GPT-6 Astra
2 of 5Language Models · verified 24 Sept 2026
GPT-6 LunaOpenAIcurrentGPT-6 AstraOpenAIcurrent| Specification | GPT-6 Luna | GPT-6 Astra |
|---|---|---|
| Summary | Ten cents a million in, for the work that does not need thinking. | Saturates three of the benchmarks it reports, and ships disabled for enterprise. |
| Context | 1.05M | 1.05M |
| Input ($/Mtok) | $0.1 | $10 |
| Output ($/Mtok) | $0.5 | $50 |
| Licence | Proprietary | Proprietary |
| Max output | 128K | 128K |
| Parameters | Not published | Not published |
| Released | 22 Sep 2026 | 3 Sep 2026 |
| Model ID | gpt-6-luna | gpt-6-astra |
| Category | Language Models | Language Models |
| Official | OpenAI ↗ | — |
Highlighted rows are where these differ.
GPT-6 Luna
- Half the price of GPT-5.6 Luna, which it replaces at the bottom of the family
- Beats or matches GPT-5.6 Luna on all five benchmarks OpenAI published
- Same 1.05M context window as the tiers above it
Best for classification and high-volume chat.
GPT-6 Astra
- 97.6% on FrontierMath Tier 4, 99.9% on ARC-AGI-3 and 100% on ExploitBench
- Above 272K input tokens the whole request reprices: input and cache at 2x, output at 1.5x
- Rolls out to OpenAI's Daybreak cyber programme first, and arrives switched off for enterprise
Best for hard reasoning and computer use.