Skip to main content
OpenAISlowUltra

o3 Pro

Highest-effort version of o3. Thinks longer with more compute to deliver the most reliable answers on the hardest science, math, and coding problems.

37,160/148,640credits
input / output ยท per 1M tokens
Highest-effort o3 reasoning
200K context window
100K max output tokens
Vision capabilities
Function calling support
Most reliable answers on hardest problems

Run it right now

Test this model instantly in the Console Playground โ€” no code required

Sign in to try

Use with AI Assistant

Copy usage instructions for Claude, ChatGPT, or other AI

Model Specifications

Context Window
200K
tokens
Max Output
100K
tokens
Training Cutoff
2024-06
Compatible SDK
OpenAI

Capabilities

Vision
Function Calling
Streaming
JSON Mode
System Prompt

Token Pricing (per 1M tokens)

Token TypeCreditsUSD Equivalent
Input Tokens37,160$24.77
Output Tokens148,640$99.09

* 1,500 credits โ‰ˆ $1 (actual charges may vary based on usage)

Quick Start

curl -X POST "https://api.core.today/llm/openai/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer cdt_your_api_key" \
  -d '{
  "model": "o3-pro",
  "messages": [
    {
      "role": "user",
      "content": "Verify whether this cryptographic key-exchange protocol is vulnerable to man-in-the-middle attacks. Enumerate every assumption and attack vector."
    }
  ],
  "max_completion_tokens": 16000
}'

Parameters

ParameterTypeRequiredDefaultDescription
messagesarrayYes-Array of message objects
modelstringYeso3-proModel identifier
max_completion_tokensintegerNo-Maximum tokens for completion (includes reasoning tokens)
reasoning_effortstringNo-Reasoning effort level: low, medium, high
lowmediumhigh

Examples

High-Stakes Verification

Most reliable reasoning with o3 Pro

curl -X POST "https://api.core.today/llm/openai/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer cdt_your_api_key" \
  -d '{
  "model": "o3-pro",
  "messages": [
    {
      "role": "user",
      "content": "Verify whether this cryptographic key-exchange protocol is vulnerable to man-in-the-middle attacks. Enumerate every assumption and attack vector."
    }
  ],
  "max_completion_tokens": 16000
}'

Tips & Best Practices

1Works with the OpenAI SDK - set base_url to https://ai.api.core.today/llm/openai/v1
2Use max_completion_tokens instead of max_tokens - temperature is not supported
3Responses can take several minutes - it thinks much longer than o3
4Reserve for problems where reliability matters most - o3 is ~10x cheaper
5Budget generous max_completion_tokens - reasoning tokens count toward the limit

Use Cases

High-stakes scientific analysis
Hardest mathematical proofs
Critical code review and verification
Complex strategic decisions