OpenAI Responses

POST /v1/responses

Use this endpoint for structured output, multimodal inputs, and newer OpenAI SDK flows.

If you are building your own GPT-6 Astra browser chat, start with the web chat integration guide.

Example

{
  "model": "gpt-4.1-mini",
  "input": "Summarize the latest Cubicspaces release notes.",
  "service_tier": "default"
}

Processing Tier

For supported models, pass service_tier as a top-level request field:

{
  "model": "gpt-6-astra",
  "input": "Summarize this material in three sentences.",
  "service_tier": "flex"
}

Supported values:

Value	Description
default	Standard processing for regular production requests
flex	Flexible processing for lower-priority, asynchronous, or batch-style work. Responses may be slower and can be unavailable when capacity is constrained
fast	Faster processing for latency-sensitive requests
priority	Compatibility alias for fast processing
auto	Use the server or project default tier

The response service_tier shows the tier that was actually used for the request. It may differ from the value sent by the client.

GPT-6 Astra Parameters

For gpt-6-astra, use max_output_tokens for the output limit. Set reasoning effort under reasoning.effort; supported values are low, medium, high, xhigh, and max.

{
  "model": "gpt-6-astra",
  "input": "Analyze this problem and use the appropriate tool.",
  "max_output_tokens": 4096,
  "reasoning": {
    "effort": "high"
  }
}

The none and minimal reasoning efforts are unsupported; start with low for ordinary chat. Do not send temperature, top_p, top_k, frequency_penalty, presence_penalty, logprobs, stop, or chat_template_kwargs. Function parameters must be valid JSON Schema, and multi-turn tool flows must return a matching function_call_output for every call_id.

Cache writes are reported as usage.input_tokens_details.cache_write_tokens.