GPT-4o - In-Depth Overview

OpenAI · GPT-4o

outdated

Flagship

Model ID: gpt-4o-2024-08-06

Get all the details on GPT-4o, an AI model from OpenAI. This page covers its token limits, pricing structure, key capabilities such as multimodal_input, long_context, function_calling, available API code samples, and performance strengths.

Key Metrics

Input Limit

128K tokens

Output Limit

16.4K tokens

Input Cost

$2.50/1M

Output Cost

$10.00/1M

Sample API Code

from openai import OpenAI
client = OpenAI()
response = client.chat.completions.create(
    model="gpt-4o-2024-08-06",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello, how are you?"}
    ]
)
print(response.choices[0].message.content)

Required Libraries

openai

Benchmarks

Benchmark	Score	Source	Notes
lmarena text	1262	OpenLLM Leaderboard	-
lmarena vision	1126	OpenLLM Leaderboard	-
lmarena search	1000	OpenLLM Leaderboard	Score for 'api-gpt-4o-search' which represents GPT-4o with search capabilities.
livebench global average	53.95	LiveBench	-
livebench reasoning average	39.75	LiveBench	-
livebench coding average	69.29	LiveBench	-
livebench mathematics average	41.48	LiveBench	-
livebench data analysis average	63.53	LiveBench	-
livebench language average	44.68	LiveBench	-
livebench if average	64.94	LiveBench	-
bfcl	72.08%	Vellum Leaderboard	-
aime 2024	13.4%	Vellum Leaderboard	-
gpqa	56.1%	Vellum Leaderboard	-
swe bench	31%	Vellum Leaderboard	-
math 500	60.3%	Vellum Leaderboard	-
alder polyglot	27.1%	Vellum Leaderboard	-

Notes

GPT-4o ('o' for 'omni') is a versatile, high-intelligence flagship model. It accepts both text and image inputs, and produces text outputs (including Structured Outputs). It is considered the best model for most tasks and is highly capable. This is a specific snapshot version of the model.