Back to catalogue

AI Inference

Structured text generation for assistants, copilots, automation, and premium user-facing experiences.

mainnet Default model: gpt-5.4Reasoning effort and output length controlsSimple paid POST API shapeMainnet only

Endpoints

1 endpoint available

Click any endpoint to see details, example requests, and try it live with your Freighter wallet.

POST /chat/respond

mainnet $0.05 USD

Text inference endpoint for prompts, system instructions, and output controls.

Example Request JSON

request.json
{
  "prompt": "Write a poem about the Stellar network.",
  "system": "Be concise and professional in your response.",
  "reasoning_effort": "medium"
}

Edit this JSON before trying the endpoint. The paid retry uses the exact same payload.

CLI Example

curl

curl -X POST "https://xlm402.com/chat/respond" -H "Content-Type: application/json" -d @- <<'JSON'
{
  "prompt": "Write a poem about the Stellar network.",
  "system": "Be concise and professional in your response.",
  "reasoning_effort": "medium"
}
JSON

Body Schema

prompt: string
system?: string
max_output_tokens?: integer 64..4096
reasoning_effort?: none | low | medium | high | xhigh
metadata?: object<string,string>