GPT-5 Mini or GPT-6 Luna: which to choose?
GPT-5 Mini (OpenAI) and GPT-6 Luna (OpenAI) compared point by point: measured performance, pricing, context and capabilities.
Data as of October 22, 2026
Side-by-side comparison
| Criterion | GPT-5 Mini | GPT-6 Luna |
|---|---|---|
| Developer | OpenAI | OpenAI |
| Released | August 7, 2025 | September 22, 2026 |
| Capability index | 62.9 / 100 | 80.3 / 100 |
| Input price | $0.25 | $0.10 |
| Output price | $2 | $0.50 |
| Cost of a typical task | $0.009 | $0.003 |
| Context window | 272,000 tokens | 1,050,000 tokens |
| Maximum output | 8,192 tokens | 128,000 tokens |
| Reasoning mode | Yes | Yes |
| Tool calling | Yes | Yes |
| Accepted inputs | text, image | text, image, PDF |
| Open weights | No | No |
| Knowledge cutoff | May 30, 2024 | May 18, 2026 |
Two models head to head
GPT-5 Mini and GPT-6 Luna are two budget models, built for high volumes at a low price, both built by OpenAI. GPT-5 Mini was released on August 7, 2025, GPT-6 Luna on September 22, 2026.
All figures below come from public sources checked on October 22, 2026: models.dev and OpenRouter for prices and capabilities, Epoch AI for scores.
Which one is more capable?
On our capability index, GPT-5 Mini scores 62.9 out of 100 and GPT-6 Luna 80.3. GPT-6 Luna leads by 17.4 points.
On the 3 benchmarks reported for both, GPT-5 Mini wins 0 times and GPT-6 Luna 3:
- GPQA Diamond (PhD-level science questions): GPT-5 Mini 75%, GPT-6 Luna 90.5%
- OTIS Mock AIME (competition mathematics problems): GPT-5 Mini 86.7%, GPT-6 Luna 98.9%
- FrontierMath (very hard research-level mathematics): GPT-5 Mini 46.7%, GPT-6 Luna 78.9%
Budget
GPT-5 Mini costs $0.25 per million input tokens and $2 per million output tokens; GPT-6 Luna costs $0.10 and $0.50.
For a typical agent task (read 20,000 tokens, write 2,000), expect $0.009 with GPT-5 Mini versus $0.003 with GPT-6 Luna. GPT-6 Luna is 3 times cheaper, a gap that adds up quickly over 1,000 tasks: $6 difference.
Run the numbers with your own volumes in the agent cost calculator.
Features and limits
GPT-6 Luna reads more in one go: 1,050,000 tokens of context (roughly 1,575 pages), versus 272,000 for GPT-5 Mini.
Reasoning mode: GPT-5 Mini yes, GPT-6 Luna yes. Tool calling: GPT-5 Mini yes, GPT-6 Luna yes. Image input: GPT-5 Mini yes, GPT-6 Luna yes.
Neither publishes its weights: both are used through an API.
Knowledge cutoff: May 30, 2024 for GPT-5 Mini, May 18, 2026 for GPT-6 Luna.
In short
Our reading of the data:
- Choose GPT-6 Luna for a higher capability index, a lower cost, a larger context window and a more recent release.
This verdict relies on public figures; it does not replace a trial on your own tasks.
Other comparisons
Frequently asked questions
GPT-5 Mini or GPT-6 Luna: which is better?
On our index (Epoch AI data), GPT-5 Mini scores 62.9 and GPT-6 Luna 80.3 out of 100. The best choice also depends on price and on your use case.
Which one is cheaper?
GPT-6 Luna. To read 20,000 tokens and write 2,000, expect $0.009 with GPT-5 Mini and $0.003 with GPT-6 Luna (prices as of October 22, 2026).
Which one has the larger context window?
GPT-6 Luna, with 1,050,000 tokens versus 272,000.
Can I run GPT-5 Mini or GPT-6 Luna on my own servers?
No: both are closed models, available through an API.
Data sources
- Models.dev — Model data: models.dev (open-source project)
- OpenRouter — Market prices cross-checked with OpenRouter
- Epoch AI — AI Benchmarking Hub — Epoch AI, “AI Benchmarking Hub”, epoch.ai, CC BY 4.0
Data as of October 22, 2026
Report an error
Does a figure, a date or a fact look wrong? Tell us: every report is checked.