# GPT-5 - Core.Today AI API

> OpenAI's latest flagship model. Delivers exceptional performance across reasoning, coding, and creative tasks with a massive 1M token context window and 32K output tokens. Supports vision, function calling, and JSON mode.

- **Provider**: OpenAI
- **Model ID**: gpt-5
- **Category**: LLM
- **Credits**: 3 per 1K tokens (avg)
- **Speed**: Fast
- **Quality**: Ultra

## Model Specifications
- **Context Window**: 1.0M tokens
- **Max Output**: 33K tokens
- **Training Cutoff**: 2025-03
- **Supported Formats**: text, json, markdown
- **Compatible SDK**: OpenAI

### Capabilities
- Vision (image input)
- Function Calling
- Streaming
- JSON Mode
- System Prompt

### Token Pricing (per 1M tokens)
- **Input Tokens**: 2,322 credits ($1.55)
- **Output Tokens**: 18,580 credits ($12.39)
- **Cached Tokens**: 232.25 credits ($0.15)

## Features
- 1M token context window
- 32K max output tokens
- Advanced reasoning capabilities
- Native vision (text + image)
- Function calling & JSON mode
- Structured outputs
- Streaming support

## Use Cases
- Complex multi-step reasoning tasks
- Advanced code generation and review
- Long document summarization and analysis
- Data extraction from documents/images
- Agentic workflows and tool use
- Enterprise content processing

## API Endpoint

Base URL: https://api.core.today
Endpoint: POST /llm/openai/v1/chat/completions

## Authentication
Header: Authorization: Bearer YOUR_API_KEY

Note: LLM endpoints use OpenAI-compatible format with Authorization Bearer token.

## Input Parameters

### Required
- **messages**: array - Array of message objects with role and content
- **model**: string (default: gpt-5) - Model identifier

### Optional
- **max_completion_tokens**: integer (default: 4096) - Maximum tokens in response (up to 32768). Note: use max_completion_tokens, not max_tokens
- **stream**: boolean (default: false) - Enable Server-Sent Events streaming
- **response_format**: object - Format of response: { type: 'json_object' } for JSON mode
- **tools**: array - List of tools (functions) the model can call
- **reasoning_effort**: string (default: medium) - Reasoning effort level: low, medium, or high

## Examples

### Complex Analysis
Multi-step reasoning and analysis with GPT-5
```json
{
  "model": "gpt-5",
  "messages": [
    {
      "role": "system",
      "content": "You are an expert analyst. Provide thorough, well-structured analysis."
    },
    {
      "role": "user",
      "content": "Analyze the potential impacts of quantum computing on current encryption standards and suggest a migration roadmap for enterprise systems."
    }
  ],
  "max_completion_tokens": 4000
}
```

### Code Generation
Generate production-quality code with GPT-5
```json
{
  "model": "gpt-5",
  "messages": [
    {
      "role": "system",
      "content": "You are an expert software engineer. Write clean, well-documented code."
    },
    {
      "role": "user",
      "content": "Create a TypeScript rate limiter middleware for Express.js that uses a sliding window algorithm with Redis for distributed rate limiting."
    }
  ],
  "max_completion_tokens": 4000
}
```

### JSON Data Extraction
Extract structured data from unstructured text
```json
{
  "model": "gpt-5",
  "messages": [
    {
      "role": "system",
      "content": "Extract structured data from the given text. Always respond with valid JSON."
    },
    {
      "role": "user",
      "content": "Parse this invoice: 'Invoice #2025-0842 from Acme Corp to Beta LLC, dated August 15 2025. Items: 10x Widget A at $25.00, 5x Widget B at $50.00. Tax: 10%. Shipping: $15.00'"
    }
  ],
  "response_format": {
    "type": "json_object"
  },
  "max_tokens": 1000
}
```

## Response Format
```json
{
  "id": "chatcmpl-abc123",
  "object": "chat.completion",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Response text here"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 100,
    "completion_tokens": 50,
    "total_tokens": 150
  }
}
```

## Tips
- Latest flagship model — best overall performance across all tasks
- 1M context window ideal for analyzing entire codebases or long documents
- Excels at complex reasoning, coding, and creative tasks
- Use lower temperature (0.1-0.3) for coding and factual tasks
- Vision works with URLs or base64-encoded images
- Streaming recommended for long responses
- Use function calling for structured tool interactions

## Documentation
https://platform.openai.com/docs/models/gpt-5