Humanlike AI
All models
OpenAI logo

GPT-5

text model · Superseded

OpenAI

The model that introduced the reasoning-effort dial.

GPT-5Idle · sample soon

Fit and price

Recruitment fit

78

Per hour

$0.499

Per minute

$0.0083

30-min interview

$0.249

About

GPT-5 was the first GPT to expose reasoning effort as a first-class parameter, and the first with a 400K context window. Superseded by 5.6 for new work, but still a sensible floor for cost-sensitive pipelines that were built against it.

  • Reasoning effort control
  • Tool use
  • Structured outputs
  • Vision

Strengths and watch-outs

Strengths

  • 400K context
  • Well-understood behaviour in existing pipelines

Watch out for

  • Superseded, avoid for new builds

Best for Pipelines pinned to GPT-5

3 variants

  • OpenAI logo

    StandardHigh

    Full model with selectable effort from minimal to high.

    5×

    rel. cost

    88

    quality

  • OpenAI logo

    MiniBalanced

    Cheaper sibling for well-specified tasks.

    1×

    rel. cost

    76

    quality

  • OpenAI logo

    NanoMini

    Cheapest rung, classification and extraction only.

    0.2×

    rel. cost

    62

    quality

Compare with