Skip to main content
OpenAISlowUltra

GPT-5 Pro

OpenAI's maximum-capability GPT-5 tier. Uses more compute to think longer and deliver the most reliable answers on the hardest reasoning, coding, and analysis tasks.

27,870/222,960credits
input / output ยท per 1M tokens
Maximum-capability GPT-5 tier
400K context window
272K max output tokens
Extended thinking for hardest problems
Native vision (text + image)
Function calling & JSON mode

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
400K
tokens
Max Output
272K
tokens
Training Cutoff
2025-03
Compatible SDK
OpenAI

Capabilities

Vision
Function Calling
Streaming
JSON Mode
System Prompt

Token Pricing (per 1M tokens)

Token TypeCreditsUSD Equivalent
Input Tokens27,870$18.58
Output Tokens222,960$148.64

* 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": "gpt-5-pro",
  "messages": [
    {
      "role": "user",
      "content": "Design a distributed consensus protocol tolerant to Byzantine faults for a 7-node cluster, prove its safety properties, and provide pseudocode."
    }
  ],
  "max_completion_tokens": 16000
}'

Parameters

ParameterTypeRequiredDefaultDescription
messagesarrayYes-Array of message objects with role and content
modelstringYesgpt-5-proModel identifier
max_completion_tokensintegerNo-Maximum tokens in response (includes internal reasoning tokens). Note: use max_completion_tokens, not max_tokens
reasoning_effortstringNo-Reasoning effort level: low, medium, or high
lowmediumhigh
streambooleanNofalseEnable Server-Sent Events streaming

Examples

Hardest Reasoning

Maximum-effort analysis with GPT-5 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": "gpt-5-pro",
  "messages": [
    {
      "role": "user",
      "content": "Design a distributed consensus protocol tolerant to Byzantine faults for a 7-node cluster, prove its safety properties, and provide pseudocode."
    }
  ],
  "max_completion_tokens": 16000
}'

Tips & Best Practices

1Works with the OpenAI SDK - set base_url to https://ai.api.core.today/llm/openai/v1
2Reserve for the hardest problems - standard GPT-5 is far cheaper for routine tasks
3Responses take longer due to extended internal thinking
4Budget generous max_completion_tokens - reasoning tokens count toward the limit

Use Cases

Hardest multi-step reasoning problems
High-stakes code review and architecture
Scientific and mathematical analysis
Complex agentic workflows
Long-document synthesis at maximum quality