GPT-4o
Accepting any input and generating any output combination of text, audio, and image for more natural interaction.
OpenAI Multimodal text, audio, and image generation
GPT-4o is a multimodal language model developed by OpenAI, released in May 2024. The "o" stands for "omni," reflecting its ability to accept any combination of text, audio, and image as input and generate any combination of those same modalities as output. It has a 128,000-token context window and a training data cutoff of October 2023.
One of GPT-4o's defining characteristics is its audio response latency, which can be as low as 232 milliseconds and averages around 320 milliseconds — comparable to human conversational response times. It is well-suited for applications requiring fast, multimodal interaction, such as voice assistants, image analysis pipelines, and multilingual text processing. OpenAI has noted it offers improved performance on non-English text compared to GPT-4 Turbo, while also being available at a lower API cost.
What GPT-4o supports
Multimodal Input
Accepts any combination of text, audio, and image inputs in a single request, enabling unified handling of mixed-media content.
Multimodal Output
Generates text, audio, and image outputs, allowing a single model to serve diverse output format requirements.
Low-Latency Audio
Responds to audio inputs in as little as 232 milliseconds, with an average response time of 320 milliseconds.
Large Context Window
Supports up to 128,000 tokens of context, enabling processing of long documents or extended conversation histories in a single call.
Multilingual Text
Handles text in a wide range of languages, with noted improvements in non-English language performance relative to GPT-4 Turbo.
Vision Understanding
Analyzes and interprets image inputs, supporting tasks such as image description, document reading, and visual question answering.
Fast Response Speed
Designed for low-latency inference, making it suitable for real-time applications and interactive user-facing products.
Cost-Effective API
Priced at approximately 50% less than GPT-4 Turbo in the API, according to OpenAI's release documentation.
Ready to build with GPT-4o?
Get Started FreeBenchmark scores
Scores represent accuracy — the percentage of questions answered correctly on each test.
| Benchmark | What it tests | Score |
|---|---|---|
| MMLU-Pro | Expert knowledge across 14 academic disciplines | 74.8% |
| GPQA Diamond | PhD-level science questions (biology, physics, chemistry) | 54.3% |
| MATH-500 | Undergraduate and competition-level math problems | 75.9% |
| AIME 2024 | American math olympiad problems | 15.0% |
| LiveCodeBench | Real-world coding tasks from recent competitions | 30.9% |
| HLE | Questions that challenge frontier models across many domains | 3.3% |
| SciCode | Scientific research coding and numerical methods | 33.3% |
Common questions about GPT-4o
What is the context window size for GPT-4o?
GPT-4o supports a context window of 128,000 tokens, which allows for long documents or extended multi-turn conversations to be processed in a single request.
What is the training data cutoff for GPT-4o?
GPT-4o has a training data cutoff of October 2023, meaning it does not have knowledge of events that occurred after that date.
What input and output types does GPT-4o support?
GPT-4o accepts any combination of text, audio, and image as input, and can generate any combination of text, audio, and image as output.
How fast does GPT-4o respond to audio inputs?
GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average response time of around 320 milliseconds, which is comparable to human conversational response times.
Is GPT-4o still available via the API?
As of February 2026, OpenAI retired GPT-4o from ChatGPT. Availability via the OpenAI API may differ; check OpenAI's official documentation for the current API model availability.
What people think about GPT-4o
Reddit discussions around GPT-4o have been notably active around its retirement from ChatGPT in February 2026, with some users expressing strong attachment to the model. A petition with approximately 20,000 signatures was organized to urge OpenAI not to remove it, and calls for subscription cancellations were reported.
Common concerns in threads center on OpenAI's decision to deprecate the model in favor of newer versions, with users debating the reasons behind the retirement. The threads reflect a user base that had integrated GPT-4o into regular workflows and was resistant to being moved to alternative models.
Why GPT 4o got retired
ChatGPT 4o is being retired today, and some users are very unhappy about it. A petition with around 20,000 signatures is urging OpenAI not to remove it. The same group is also calling for a mass cancellation of subscriptions in protest.
OpenAI will retire GPT-4o, GPT-4.1, GPT-4.1 mini, and o4-mini from ChatGPT on February 13
Parameters & options
Explore similar models
OpenAI GPT‑5.2 Pro
The most capable model series yet for professional knowledge work.
OpenAI GPT-5.1
The best model for coding and agentic tasks across domains
OpenAI GPT-5
The best model for coding and agentic tasks across domains
Google Gemini 3.1 Pro
Google's frontier reasoning model delivering enhanced software engineering, agentic reliability, and multimodal intelligence across a 1M-token context window.
Amazon Amazon Nova Lite
Very low cost multimodal model that is lightning fast for processing image, video, and text inputs.
Nvidia Nemotron 3 Nano 30B
NVIDIA's compact open-weight hybrid MoE model with 30B total / 3.5B active parameters, delivering strong reasoning and coding performance up to 1M context.
Start building with GPT-4o
No API keys required. Create AI-powered workflows with GPT-4o in minutes — free.