o1-pro
OpenAI's most powerful reasoning model, built to tackle the hardest problems with extended thinking time and enhanced compute.
OpenAI Extended reasoning for complex, high-stakes problems
o1-pro is a text generation model developed by OpenAI and released in December 2024. It is built on the same foundation as the o1 model family but allocates significantly more compute and longer reflection time per query, which allows it to work through multi-step problems more carefully before producing a response. It supports a 200,000-token context window and can generate up to 100,000 tokens in a single output, and it accepts both text and image inputs.
The model is designed for tasks where accuracy on difficult problems takes priority over response speed. It performs well on advanced mathematics, scientific reasoning, and complex coding challenges, with benchmark scores including 94.8% on MATH, 92.4% on HumanEval, and 77.3% on GPQA. o1-pro was initially available exclusively through the ChatGPT Pro subscription plan before becoming accessible via the OpenAI API in March 2025.
What o1-pro supports
Extended Reasoning
The model spends additional compute time reflecting before responding, allowing it to work through multi-step problems and catch errors that faster inference would miss.
Large Context Window
Supports up to 200,000 tokens of input, equivalent to roughly 300 pages of text, making it suitable for long documents and large codebases.
High Output Capacity
Can generate up to 100,000 tokens in a single response, enabling detailed, long-form answers to complex queries.
Image Input
Accepts image inputs alongside text prompts, outputting detailed text responses based on visual and written context.
Advanced Math Solving
Achieves 94.8% pass@1 on the MATH benchmark, reflecting strong performance on graduate-level and competition mathematics.
Code Generation
Scores 92.4% on HumanEval, demonstrating reliable ability to write, analyze, and debug code across programming tasks.
Scientific Reasoning
Scores 77.3% on GPQA, a benchmark of graduate-level questions in biology, chemistry, and physics.
Ready to build with o1-pro?
Get Started FreeCommon questions about o1-pro
What is the context window for o1-pro?
o1-pro supports a context window of 200,000 tokens for input, which is roughly equivalent to 300 pages of text.
How much does o1-pro cost to use via the API?
As of its API release in March 2025, o1-pro is priced at $150 per million input tokens and $600 per million output tokens, making it one of the higher-priced models in OpenAI's API catalog.
What is the knowledge cutoff date for o1-pro?
o1-pro was released in December 2024. OpenAI has indicated the o1 model family has a knowledge cutoff of October 2023.
How does o1-pro differ from the standard o1 model?
o1-pro uses more compute and longer reflection time per query compared to the standard o1 model. This is intended to produce more accurate and reliable answers on complex tasks, at the cost of slower response times and higher pricing.
What types of inputs does o1-pro accept?
o1-pro accepts both text and image inputs and returns text outputs. It does not generate images or audio.
What people think about o1-pro
Community discussion around o1-pro has focused largely on its pricing and the timing of its API release, with many users expressing surprise that OpenAI made it available via API in March 2025 alongside GPT-4.5. Some users in the LocalLLaMA subreddit found the decision unusual given the model's high cost relative to other available options.
A recurring concern across threads is whether the pricing of o1-pro and similar flagship models reflects a deliberate strategy rather than standard market positioning. Practical use cases discussed include high-stakes reasoning tasks where accuracy is critical, though many users noted the cost makes it impractical for general or high-volume applications.
OpenAI released GPT-4.5 and O1 Pro via their API and it looks like a weird decision.
With the insane prices of recent flagship models like GPT-4.5 and O1-Pro, is OpenAI trying to limit DeepSeek's use of its API for training?
Documentation & links
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.
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.
X.ai Grok 4.20 Reasoning
A beta reasoning variant of xAI's Grok 4.20, designed for enhanced step-by-step thinking and complex problem-solving.
Start building with o1-pro
No API keys required. Create AI-powered workflows with o1-pro in minutes — free.