OpenAI GPT-4o API

openai

gpt-4o

Warm

OpenAI's high-intelligence chat model

Official

170.9K runs

Priced by multiple properties

Commercial use

Run with an API

Pricing

License

Playground API Examples README

Input

prompt

string

Shift + Return to add a new line

Who was the 16th president of the United States?Who was the 16th president of the United States?

The prompt to send to the model. Do not use if using messages.

system_prompt

string

Shift + Return to add a new line

System prompt to set the assistant's behavior

messages

A JSON string representing a list of messages. For example: [{"role": "user", "content": "Hello, how are you?"}]. If provided, prompt and system_prompt are ignored.

Default: []

image_input

file[]

Add multiple files

List of images to send to the model

Default: []

temperature

number

(minimum: 0, maximum: 2)

Sampling temperature between 0 and 2

Default: 1

max_completion_tokens

integer

Maximum number of completion tokens to generate

Default: 4096

top_p

number

(minimum: 0, maximum: 1)

Nucleus sampling parameter - the model considers the results of the tokens with top_p probability mass. (0.1 means only the tokens comprising the top 10% probability mass are considered.)

Default: 1

frequency_penalty

number

(minimum: -2, maximum: 2)

Frequency penalty parameter - positive values penalize the repetition of tokens.

Default: 0

presence_penalty

number

(minimum: -2, maximum: 2)

Presence penalty parameter - positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.

Default: 0

Run this model in Node.js with one line of code:

npx create-replicate --model=openai/gpt-4o

or set up a project from scratch

Install Replicate’s Node.js client library:

npm install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run openai/gpt-4o using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const input = {
  top_p: 1,
  prompt: "Who was the 16th president of the United States?",
  messages: [],
  image_input: [],
  temperature: 1,
  system_prompt: "You are a helpful assistant.",
  presence_penalty: 0,
  frequency_penalty: 0,
  max_completion_tokens: 4096
};

for await (const event of replicate.stream("openai/gpt-4o", { input })) {
  process.stdout.write(event.toString());
};

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Import the client:

import replicate

Run openai/gpt-4o using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

# The openai/gpt-4o model can stream output as it's running.
for event in replicate.stream(
    "openai/gpt-4o",
    input={
        "top_p": 1,
        "prompt": "Who was the 16th president of the United States?",
        "messages": [],
        "image_input": [],
        "temperature": 1,
        "system_prompt": "You are a helpful assistant.",
        "presence_penalty": 0,
        "frequency_penalty": 0,
        "max_completion_tokens": 4096
    },
):
    print(str(event), end="")

To learn more, take a look at the guide on getting started with Python.

Set the REPLICATE_API_TOKEN environment variable:

export REPLICATE_API_TOKEN=<paste-your-token-here>

Find your API token in your account settings.

Run openai/gpt-4o using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "input": {
      "top_p": 1,
      "prompt": "Who was the 16th president of the United States?",
      "messages": [],
      "image_input": [],
      "temperature": 1,
      "system_prompt": "You are a helpful assistant.",
      "presence_penalty": 0,
      "frequency_penalty": 0,
      "max_completion_tokens": 4096
    }
  }' \
  https://api.replicate.com/v1/models/openai/gpt-4o/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

The 16th president of the United States was Abraham Lincoln. He served from March 4, 1861, until his assassination on April 15, 1865.

Generated in

1.1 seconds

Input tokens

Output tokens

Tokens per second

59.70 tokens / second

Time to first token

0.6 seconds

Tweak it Iterate in playground Report View full prediction

Pricing

Model pricing for openai/gpt-4o. Looking for volume pricing? Get in touch.

$0.01

per thousand output tokens

or 100,000 tokens for $1

$2.50

per million input tokens

or 400,000 tokens for $1

Official models are always on, maintained, and have predictable pricing. Learn more.

Check out our docs for more information about how pricing works on Replicate.

Readme

GPT‑4o is OpenAI’s most advanced flagship model, offering natively multimodal capabilities across text, vision, and audio. It delivers GPT-4‑level performance with faster response times and lower cost, making it ideal for real-time, high-volume applications. GPT‑4o supports audio inputs and outputs, handles images and text simultaneously, and is designed to feel conversational and responsive — like interacting with a human assistant in real time.

Key Capabilities

Multimodal input & output: Supports text, images, audio (input) and audio/text (output)
Real-time audio responsiveness: Latency as low as 232 ms
1M token context window for deep reasoning over long content (API)
High performance across reasoning, math, and code tasks
Unified model for all modalities—no need to switch between specialized models

Benchmark Highlights

MMLU (Language understanding):        87.2%
HumanEval (Python coding):           90.2%
GSM8K (Math word problems):          94.4%
MMMU (Vision QA):                    74.1%
VoxCeleb (Speaker ID):               95%+ (est.)
Audio Latency (end-to-end):          ~232–320ms

Use Cases

Real-time voice assistants and spoken dialogue agents
Multimodal document Q&A (PDFs with diagrams, charts, or images)
Code writing, explanation, and debugging
High-volume summarization and extraction from audio/text/image
Tutoring, presentations, and interactive education tools

🔧 Developer Notes

Available via OpenAI API and ChatGPT (Free, Plus, Team, Enterprise)
In ChatGPT, GPT‑4o is now the default GPT-4-level model
Audio input/output is supported only in ChatGPT for now
Image and text input supported via both API and ChatGPT
Supports streaming, function calling, tool use, and vision APIs
Context window of 128k tokens in ChatGPT; 1M tokens via API (limited release)