ZMIME
OpenAI · Released May 13, 2024

GPT-4o

Omni-era GPT for multimodal chat, practical coding, and general assistants.

  • Proprietary
  • Tool calling
  • Vision
Capability
129.0ECI · #129 of 148
Input
$2.50per 1M tokens
Output
$10.00per 1M tokens
Context
128K16K max output
Pricing

What it costs

Input
$2.50per million tokens
Output
$10.00per million tokens
Cached input
$1.25per million tokens
Blended (3:1)
$4.38pricier than 84% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$10.00
Team assistant25M input + 5M output tokens$112.50
Production app250M input + 50M output tokens$1,125

Official OpenAI API price, as listed on models.dev.

Capability

Independent benchmarks

129.0Epoch Capabilities Index
#129 of 148 scored models
88Median 146167

The shaded band is Epoch AI’s confidence range (124.2–131.5); the tick marks the median scored model.

  • GPQA DiamondGraduate-level science questions48.9%
  • OTIS Mock AIME 2024–2025Competition mathematics6.3%

Scores from Epoch AI, run independently of OpenAI.

Specifications

The details

Lab
OpenAI
Released
May 13, 2024
Knowledge cutoff
Sep 2023
Context window
128,000 tokens
Max output
16,384 tokens
Inputs
Text, Images, PDFs
Output
Text
Reasoning
No
Tool calling
Yes
Structured output
Yes
Weights
Proprietary
API model ID
gpt-4oon OpenAI
Availability
19 API providerslisted on models.dev

OpenAI model documentation

Lab-reported

What OpenAI claims

Published by the lab at launch. Settings vary, so compare these only with care.

BenchmarkScoreSettingSource
Aider Polyglot23.1 percent correct—Source
Same lab

More from OpenAI

Questions

About GPT-4o

How much does GPT-4o cost?

GPT-4o costs $2.50 input / $10.00 output per million tokens (official OpenAI API price). Cached input is $1.25 per million tokens. At a 3:1 input-to-output mix that is $4.38 per million tokens, more expensive than 84% of the 360 priced models we track.

What is the context window of GPT-4o?

GPT-4o accepts up to 128,000 tokens per request and can write up to 16,384 tokens in one response.

How good is GPT-4o?

Epoch AI gives GPT-4o a Capabilities Index score of 129.0 (likely range 124.2–131.5), ranking it #129 of 148 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 48.9%, OTIS Mock AIME 2024–2025 6.3%.

Is GPT-4o open source?

No. GPT-4o is proprietary; you use it through OpenAI’s API or partner platforms.

What inputs does GPT-4o support?

GPT-4o accepts text, images and PDFs and replies in text. It is not a dedicated reasoning model, supports tool calling and can return structured JSON output.

When was GPT-4o released?

OpenAI released GPT-4o on May 13, 2024. Its training data runs to Sep 2023.

What are the best alternatives to GPT-4o?

The closest current models from other labs on capability, price and release date are Qwen2.5 72B Instruct (Alibaba (Qwen), ECI 129.0, $1.40 / $5.60), Mistral Large 2.1 (Mistral AI, ECI 128.5, $2.00 / $6.00), Claude Sonnet 3.5 v2 (Anthropic, ECI 133.5, $3.00 / $15.00) and Llama-3.1-70B-Instruct (Meta, ECI 125.9, $0.72 / $0.72).