GPT-4o
GPT-4o is an omni-modal model from OpenAI that accepts text, audio, and image inputs and generates text, audio, and image outputs.
Omni-modal text, audio, and image generation
GPT-4o, where the "o" stands for "omni," is a model developed by OpenAI and released in May 2024. It is designed to accept any combination of text, audio, and image as input and produce any combination of text, audio, and image as output, enabling more natural human-computer interaction. The model 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.
GPT-4o is well-suited for tasks that benefit from multimodal understanding, including vision-based tasks, multilingual text processing, and real-time conversational applications. It was noted at launch to match GPT-4 Turbo performance on English text and code while offering meaningful improvements on non-English languages. The model carries a 128,000-token context window and supports a maximum response size of 16,384 tokens, making it applicable to longer document processing and extended dialogue tasks.
What GPT-4o supports
Multimodal Input
Accepts any combination of text, audio, and image inputs in a single request, enabling unified processing across modalities.
Multimodal Output
Generates outputs in text, audio, or image format, allowing flexible response types depending on the use case.
Low-Latency Audio Response
Responds to audio inputs with an average latency of 320 milliseconds, with a minimum of 232 milliseconds, approximating human conversational timing.
Large Context Window
Supports a 128,000-token context window, allowing processing of long documents, extended conversations, or large code files in a single request.
Multilingual Text Processing
Handles text in a wide range of languages, with noted improvements over prior models on non-English language tasks.
Cost-Effective API Access
Priced at 50% less than GPT-4 Turbo in the API at launch, reducing cost for high-volume or production workloads.
Vision Understanding
Analyzes and reasons about image inputs, supporting tasks such as image description, visual question answering, and document parsing.
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 has a context window of 128,000 tokens, meaning it can process up to 128,000 tokens of combined input and conversation history in a single request.
What is the maximum response size for GPT-4o?
GPT-4o supports a maximum response size of 16,384 tokens per completion.
What input types does GPT-4o support?
GPT-4o is designed to accept any combination of text, audio, and image inputs, making it an omni-modal model rather than a text-only one.
Is GPT-4o still actively supported?
GPT-4o currently has a status of deprecated in the MindStudio catalog. OpenAI has released newer model versions, so developers should check OpenAI's documentation for the recommended current model.
Who publishes GPT-4o and where is it hosted?
GPT-4o is published by OpenAI and is provided as a first-party model, meaning it is accessed directly through OpenAI's infrastructure without a third-party intermediary.
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
OpenAIGPT-5.5
GPT-5.5 is a multimodal text generation model from OpenAI with a 1,000,000-token context window and adjustable reasoning effort.
OpenAIGPT OSS 120B
GPT OSS 120B is an open-source text generation model from OpenAI, served via Groq with a 128,000-token context window.
OpenAIGPT OSS 20B
GPT OSS 20B is an open-source text generation model from OpenAI with a 128,000-token context window, served via Groq.
GoogleGemma 4 26B
Gemma 4 26B is a multimodal text generation model from Google with a 262,144 token context window.
MetaMuse Glimmer 30B
Muse Glimmer 30B is a multimodal text generation model from Meta with a 131,072 token context window.
QwenQwen3.8 27B
Start building with GPT-4o
No API keys required. Create AI-powered workflows with GPT-4o in minutes — free.