The Article

GPT-5: Key characteristics, pricing and model card

Published on August 7, 2025

An overview of the new GPT-5 model family, covering its key characteristics, competitive pricing structure, and insights from the official OpenAI system card, based on two weeks of preview access.

Key Sections from the Article

Key model characteristics

GPT-5 is a hybrid system with smart, reasoning, and mini models. API versions have four reasoning levels and high token limits.

Position in the OpenAI model family

The three new GPT-5 models are intended to replace most of the existing OpenAI line-up, except for audio and image generation.

Pricing is aggressively competitive

GPT-5's pricing is highly competitive, with the main model at half the input cost of GPT-4o and significant caching discounts.

Prompt injection in the system card

While improved, with a 56.8% attack success rate, prompt injection remains an unsolved problem for GPT-5 according to red-teaming assessments.

Software & Offers

GPT-5

Large Language Model

by OpenAI

Pricing

$1.25 / 1M input tokens

$10.00 / 1M output tokens

GPT-5 Mini

Large Language Model

by OpenAI

Pricing

$0.25 / 1M input tokens

$2.00 / 1M output tokens

GPT-5 Nano

Large Language Model

by OpenAI

Pricing

$0.05 / 1M input tokens

$0.40 / 1M output tokens

How-Tos

How to Access GPT-5 Thinking Traces via the API

A step-by-step guide to retrieve reasoning traces from the GPT-5 API using a curl command.

  1. Construct the API Request: Begin by constructing a curl command targeting the `https://api.openai.com/v1/responses` endpoint.
  2. Add Headers: Include the necessary headers: `Authorization: Bearer YOUR_API_KEY` and `Content-Type: application/json`.
  3. Define the Request Body: In the data payload, specify the model, your input, and the crucial reasoning parameter: `'reasoning': {'summary': 'auto'}`.
  4. Execute and Inspect: Execute the command. The JSON response will contain a 'reasoning' field with the model's thinking traces.

How to Choose the Right GPT-5 Model

A guide on selecting between gpt-5, gpt-5-mini, and gpt-5-nano based on cost, performance, and task complexity.

  • For Max Performance: Choose the main GPT-5 model for the highest competence and complex reasoning.
  • For Balance: Select GPT-5 Mini for a good balance of cost and performance.
  • For Speed & Low Cost: Use GPT-5 Nano for the fastest responses and lowest cost on simpler tasks.

Key Terminology

Safe-Completions

A safety-training approach that centers on the safety of the assistant’s output rather than a binary classification of the user’s intent.

Sycophancy

The tendency of a model to produce outputs that align with a user's perceived preferences or biases, which OpenAI has post-trained GPT-5 to reduce.

Thinking Traces

The internal reasoning steps a model takes to generate a final response, which can be exposed via the API using a specific parameter.

Prompt Injection

An attack technique where malicious input is crafted to make an LLM ignore its original instructions and perform unintended actions.

Reasoning Effort

A parameter for GPT-5 API models that controls the amount of computational 'thinking' the model performs before generating a response.

System Card

A document from AI developers detailing a model's capabilities, limitations, training, and safety evaluations.

Frequently Asked Questions

GPT-5 is a unified system from OpenAI with a smart model for most questions, a deeper reasoning model for harder problems, and a real-time router to select the appropriate model.

The models have an input limit of 272,000 tokens and an output limit of 128,000 tokens, which includes invisible reasoning tokens.

There is a significant 90% discount on input tokens that have been used within the previous few minutes, which is beneficial for chat applications.

People & Organizations

Simon Willison

Software developer, journalist, and creator of Simon Willison's Weblog.

Visit Website

OpenAI

An AI research and deployment company. Our mission is to ensure that artificial general intelligence benefits all of humanity.

Visit Website