o3 Pro
o3 Pro applies maximum reasoning compute to the o3 architecture, delivering the deepest chain-of-thought analysis available for the most demanding mathematical, scientific, and analytical problems.
import { streamText } from 'ai'
const result = streamText({ model: 'openai/o3-pro', prompt: 'Why is the sky blue?'})Playground
Try out o3 Pro by OpenAI. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Providers
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
More models by OpenAI
| Model |
|---|
| 1M | 0.9s | 71tps | $5.00/M | $30.00/M | Read: $0.5/M Write: — | $10.00/K + input costs | — |   | 04/24/2026 | ||||
| 400K | 1.3s | 70tps | $0.75/M | $4.50/M | Read:$0.07/M Write:— | $10.00/K + input costs | — | | 03/17/2026 | ||||
| 400K | 0.6s | 25tps | $0.20/M | $1.25/M | Read:$0.02/M Write:— | $10.00/K + input costs | — | | 03/17/2026 | ||||
| 1.1M | 0.6s | 56tps | $2.50/M | $15.00/M | Read: $0.25/M Write: — | $10.00/K + input costs | — | | 03/05/2026 | ||||
| 128K | 0.6s | 105tps | $1.25/M | $10.00/M | Read:$0.13/M Write:— | $10.00/K + input costs | — | | 11/12/2025 | ||||
| 131K | 0.1s | 210tps | $0.35/M | $0.75/M | Read:$0.25/M Write:— | — | — |    +5 | 08/05/2025 |
About o3 Pro
o3 Pro was released on April 16, 2025 as the premium tier of the o3 reasoning model family. It applies maximum compute to each reasoning request, enabling the deepest chain-of-thought analysis available from OpenAI.
The additional compute produces more thorough deliberation: the model explores more solution paths, checks its work more exhaustively, and arrives at conclusions with higher confidence. On the hardest mathematical, scientific, and analytical problems, this translates to measurably better accuracy.
The tradeoff is cost and latency. o3 Pro is the most expensive model in the o3 family and takes the longest to respond. For problems where the difference between good and best matters, that investment is justified. For standard reasoning tasks, o3 or o3-mini provide sufficient quality at substantially lower cost.
What To Consider When Choosing a Provider
- Configuration: o3 Pro applies the most compute per request of any o3 variant. Use it when the problem genuinely requires exhaustive deliberation and the quality difference is consequential.
- Configuration: Most teams reserve o3 Pro for their hardest queries, routing standard reasoning traffic to o3 or o3-mini.
- Zero Data Retention: AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
When to Use o3 Pro
Best For
- Frontier mathematical reasoning: The hardest proofs, competition problems, and mathematical research
- Critical scientific analysis: Complex derivations and multi-step scientific reasoning where errors have consequences
- Exhaustive code analysis: Security audits and algorithmic verification requiring thorough deliberation
- Research-grade reasoning: Academic and research problems that benefit from the deepest available analysis
- High-stakes analytical decisions: Analyses where maximum reasoning depth directly impacts outcomes
Consider Alternatives When
- Standard reasoning: O3 for most reasoning tasks at lower cost
- Budget-friendly reasoning: O3-mini for cost-efficient chain-of-thought on routine problems
- General-purpose tasks: GPT-5.2 pro for tasks that need premium quality but not specialized reasoning
- Fast responses: Any standard model when response time matters more than reasoning depth
Conclusion
o3 Pro applies maximum compute within the o3 family to the hardest problems. For mathematical, scientific, and analytical challenges where reasoning depth directly determines quality, it is the premium tier of the o3 lineup.
Frequently Asked Questions
How does o3 Pro differ from standard o3?
It applies more compute per request, enabling deeper and more exhaustive chain-of-thought reasoning. The improvement is most significant on the hardest problems.
When is the premium cost justified?
For frontier mathematical proofs, critical scientific analysis, security audits, and other problems where the deepest possible reasoning produces meaningfully better outcomes.
What context window does o3 Pro support?
200K tokens, matching the o3 family.
Is o3 Pro slower than o3?
Yes. The additional reasoning compute increases response time. Reserve it for problems where depth justifies the wait.
How does AI Gateway handle authentication for o3 Pro?
AI Gateway accepts a single API key or OIDC token for all requests. You don't embed OpenAI credentials in your application; AI Gateway routes and authenticates on your behalf.
What are typical latency characteristics?
This page shows live throughput and time-to-first-token metrics measured across real AI Gateway traffic.