← All benchmarks

B-Tier

GPT-6 Luna

OpenAI · Lightweight everyday model

OpenAI's lightest current-generation model — built for volume, not depth.

10 of 10 benchmark runs completed

Why this tier

Luna is the entry point to the GPT-6 family, and it performs exactly like one: ten consistent, serviceable runs on everyday tasks, with a clear and expected drop-off once a task needs real reasoning depth. That's not a criticism — it's priced and positioned for high-volume, low-complexity work, and it's honest B-tier for that job.

Strengths

  • Fast and inexpensive relative to the rest of the GPT-6 family
  • Consistent, dependable performance on everyday tasks
  • Easiest on-ramp into the OpenAI ecosystem

Worth knowing

  • Clear capability ceiling on complex reasoning or agentic work
  • Will need escalation to Sol or Astra for anything genuinely hard

Best for

  • High-volume, low-complexity support and assistant tasks
  • Draft generation and first-pass content
  • Cost-sensitive products at scale

Not sure which model fits?

We'll help you pick the right one for the job.

Model choice is rarely one-size-fits-all. Tell us what you're building and we'll tell you plainly what fits.