GPT 5.5

model: 'gpt-5-5'

Available through openai and openrouter.

Configuration

// OpenAI
{
  provider: 'openai',
  model: 'gpt-5-5',
  extra: { apiKey: process.env.OPENAI_API_KEY },
}

// OpenRouter
{
  provider: 'openrouter',
  model: 'gpt-5-5',
  extra: { apiKey: process.env.OPENROUTER_API_KEY },
}

See OpenAI or OpenRouter for authentication.

Specifications

Max input tokens	1,000,000
Max output tokens	128,000
Temperature	Not sent — sampling parameters are unsupported on this model

Cost

	Cost
Input tokens	$5.00 / million
Output tokens	$30.00 / million

Costs are estimates. Actual billing depends on the provider and region.

JSON output

JSON enrichers use response_format: json_schema (strict: true). See JSON output and schemas.

Timeout

timeoutMs applies per individual LLM call within a batch. If unset, no explicit timeout is applied beyond the provider's default.

Related

• OpenAI
• LLM models overview