o3
OpenAI's reasoning model following o1, released April 2025, and the first of its reasoning models to combine tools agentically.
Key Specifications
- Context window
- 200K tokens
- Max output
- 100K tokens
- Knowledge cutoff
- web search available
- Released
- Inputs
- Text, image
- Capabilities Show details
- Reasoning Thinks before it answers, always on or as a switchable mode. Function calling Connect to external tools, APIs, and systems. Structured output Return responses in structured formats like JSON. Web search Search the web for up-to-date information. Code execution Write and execute code in a sandboxed environment. File upload Process uploaded documents, spreadsheets, and other files. MCP Connect to services via Model Context Protocol. Image generation Generate and edit images from text prompts.
Hosted API pricing
| Provider | Input /1M tokens | Output /1M tokens | Cached input /1M | Cost at 10M in + 2M out | |
|---|---|---|---|---|---|
|
OpenAI
Creator
|
$2.00 | $8.00 | $0.50 | $36.00 | View |
DigitalOcean
|
$2.00 | $8.00 | $0.50 | $36.00 | View |
Azure
|
$2.00 | $8.00 | $0.50 | $36.00 | View |
Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.
Similarly priced models
The models nearest o3 by blended rate, each at its own cheapest provider.
| Model | Blended /1M | Input /1M | Output /1M | Context | Cutoff | vs o3 |
|---|---|---|---|---|---|---|
Gemini 2.5 Pro
Google Cloud
|
$2.7083 | $1.25 | $10.00 | 1M | Jan 2025 | −10% |
|
GPT-5
OpenAI
|
$2.7083 | $1.25 | $10.00 | 400K | Sep 2024 | −10% |
|
GPT-5.1
OpenAI
|
$2.7083 | $1.25 | $10.00 | 400K | Sep 2024 | −10% |
|
Gemini 3.5 Flash
Google Cloud
|
$2.75 | $1.50 | $9.00 | 1M | −8% | |
|
GPT-4.1
OpenAI
|
$3.00 | $2.00 | $8.00 | 1M | Jun 2024 | 0% |
|
o3
This model
OpenAI
|
$3.00 | $2.00 | $8.00 | 200K | Jun 2024 | |
Claude Sonnet 5
Anthropic
|
$3.3333 | $2.00 | $10.00 | 1M | Jan 2026 | +11% |
|
GPT-5.6 Sol
OpenAI
|
$3.3333 | $2.00 | $10.00 | 1M | Feb 2026 | +11% |
|
Gemini 3.1 Pro Preview
Google Cloud
|
$3.6667 | $2.00 | $12.00 | 1M | Jan 2025 | +22% |
|
GPT-5.6 Terra
OpenAI
|
$3.6667 | $2.00 | $12.00 | 1M | Feb 2026 | +22% |
Kimi K3
Moonshot AI
|
$3.6667 | $2.00 | $12.00 | 1M | +22% |
Prices are USD per 1M tokens at each model's cheapest listed provider. Blended is the cost of 10M input plus 2M output tokens, spread over the 12M.
Frequently Asked Questions
What is o3 good for?
Multi-step problems that mix text, code and images with tool use. OpenAI calls it the first of its reasoning models to combine tools agentically, mid-priced in the lineup.
When is o3 not a good fit?
Long-lived deployments: OpenAI says GPT-5 succeeds it and plans to shut it down in December 2026. Its knowledge is older and its inputs shorter than the GPT-5 line's.
What is the cheapest way to run o3?
The cheapest hosted rate is $2.00 / $8.00 per 1M tokens via OpenAI.
How does o3 compare with o1?
OpenAI says o3 is a step up from o1 in tool use: it is trained to call tools inside its chain of thought, deciding when and how to use them, and can work with images as part of its reasoning, cropping or rotating them as it thinks. Same 200K-token context and text and image input. o3 lists at 87% less per 1M input tokens and 87% less per 1M output tokens than o1. It came out 4 months later and has a newer knowledge cutoff (June 2024 against October 2023).
Can I self-host o3?
No. OpenAI does not publish the weights; o3 is available only through hosted APIs.
More from OpenAI
| Model | Context | Input /1M | Output /1M |
|---|---|---|---|
|
GPT-4.1
|
1M | $2.00 | $8.00 |
|
GPT-5
|
400K | $1.25 | $10.00 |
|
GPT-5.1
|
400K | $1.25 | $10.00 |
|
GPT-5.6 Sol
|
1M | $2.00 | $10.00 |
|
GPT-5.6 Terra
|
1M | $2.00 | $12.00 |
|
GPT-4o
|
128K | $2.50 | $10.00 |
|
GPT-5.2
|
400K | $1.75 | $14.00 |
|
GPT-5.3-Codex
|
400K | $1.75 | $14.00 |