Question 1

What is o3 Pro?

Accepted Answer

o3 Pro is a model from OpenAI, released on Jun 10, 2025. The o-series of models are trained with reinforcement learning to think before they answer and perform complex reasoning. The o3-pro model uses more compute to think harder and provide consistently better answers. Note that BYOK is required for this model. Set up here: https://openrouter.ai/settings/integrations.

Question 2

How much does o3 Pro cost?

Accepted Answer

On the Respan gateway, input starts at $20.00 per 1M tokens on OpenRouter, and output starts at $80.00 per 1M tokens on OpenRouter, after gateway discounts.

Question 3

Which providers serve o3 Pro on Respan?

Accepted Answer

o3 Pro is served by OpenRouter.

Question 4

What is o3 Pro's context window?

Accepted Answer

o3 Pro accepts up to 200K tokens of context.

Question 5

How do I use o3 Pro with the Respan gateway?

Accepted Answer

Point any OpenAI-compatible SDK at https://api.respan.ai/api/ with your Respan API key and set the model to "o3-pro". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Question 6

How fast is o3 Pro?

Accepted Answer

Over the last 30 days of requests through the Respan gateway, the fastest provider was OpenRouter at 7 tokens per second, and the lowest response latency was 3.9 seconds on OpenRouter. Speed varies with request size and provider.

Question 7

How does o3 Pro score on benchmarks?

Accepted Answer

o3 Pro has 1 published benchmark score, including WeirdML 58.2%. Each score links to where it was published; the data is collected by llmmetric.com.

o3 Pro

Providers

Pricing

Cost by volume

Throughput and latency

Output speed

Response latency

Certified where it counts.

Benchmarks

Coding1

Apps

o3 Pro FAQ

Works with your entire stack

Built for AI agents.
Break less.
Ship more.