GPT-5.6 Luna vs GPT-5 Mini

The match-up at a glance

GPT-5.6 Luna and GPT-5 Mini are two budget models, built for high volumes at a low price, both built by OpenAI. GPT-5.6 Luna was released on July 9, 2026, GPT-5 Mini on August 7, 2025.

All figures below come from public sources checked on October 22, 2026: models.dev and OpenRouter for prices and capabilities, Epoch AI for scores.

Performance

On our capability index, GPT-5.6 Luna scores 79.6 out of 100 and GPT-5 Mini 62.9. GPT-5.6 Luna leads by 16.7 points.

On the 3 benchmarks reported for both, GPT-5.6 Luna wins 3 times and GPT-5 Mini 0:

• GPQA Diamond (PhD-level science questions): GPT-5.6 Luna 91.6%, GPT-5 Mini 75%
• OTIS Mock AIME (competition mathematics problems): GPT-5.6 Luna 98.3%, GPT-5 Mini 86.7%
• FrontierMath (very hard research-level mathematics): GPT-5.6 Luna 82.1%, GPT-5 Mini 46.7%

Pricing

GPT-5.6 Luna costs $0.20 per million input tokens and $1.2 per million output tokens; GPT-5 Mini costs $0.25 and $2.

For a typical agent task (read 20,000 tokens, write 2,000), expect $0.006 with GPT-5.6 Luna versus $0.009 with GPT-5 Mini. GPT-5.6 Luna is 1.4 times cheaper, a gap that adds up quickly over 1,000 tasks: $2.6 difference.

Run the numbers with your own volumes in the agent cost calculator.

Under the hood

GPT-5.6 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-5.6 Luna yes, GPT-5 Mini yes. Tool calling: GPT-5.6 Luna yes, GPT-5 Mini yes. Image input: GPT-5.6 Luna yes, GPT-5 Mini yes.

Neither publishes its weights: both are used through an API.

Knowledge cutoff: February 16, 2026 for GPT-5.6 Luna, May 30, 2024 for GPT-5 Mini.

In short

Our reading of the data:

• Choose GPT-5.6 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.