DocsAPI referenceResponses API

Responses API

The OpenAI-compatible Responses endpoint — request fields, response shape, streaming.

POST/v1/responses

A chat endpoint compatible with the OpenAI Responses API request body — the same interface coding tools like Codex use. Authenticate with Authorization: Bearer.

Request body

Field	Type	Required	Description
`model`	string	Yes	Model id — see Models.
`input`	string \| array	Yes	User input: plain text, or an array of messages for multi-turn / multimodal input.
`instructions`	string	No	System prompt.
`max_output_tokens`	integer	No	Output token cap (minimum 16).
`stream`	boolean	No	When `true`, the response streams over SSE — see Streaming.
`temperature` / `tools` / …	—	No	Other official fields are forwarded as-is; support depends on the target model.

request.sh

curl https://soleapi.com/v1/responses \
  -H "Authorization: Bearer $SOLEAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-fable-5",
    "input": "Write a haiku about the sea",
    "max_output_tokens": 200
  }'

Response

response.json

{
  "id": "resp_abc123",
  "object": "response",
  "model": "claude-fable-5",
  "status": "completed",
  "output": [
    {
      "type": "message",
      "role": "assistant",
      "content": [
        { "type": "output_text", "text": "Waves fold into foam..." }
      ]
    }
  ],
  "usage": {
    "input_tokens": 14,
    "output_tokens": 23,
    "total_tokens": 37
  }
}

The token counts in usage are what billing is based on; official SDKs expose the plain text via resp.output_text.

Other OpenAI-compatible endpoints

The same auth header also covers the endpoints below; request and response shapes match the OpenAI originals:

Endpoint	Purpose	Billing
`POST /v1/embeddings`	Text embeddings (`text-embedding-3-small` / `-large`)	Per input token
`POST /v1/images/generations` · `/v1/images/edits`	Image generation and editing, see Images & video	Per image
`POST /v1/videos/generations` · `/v1/videos/edits` · `GET /v1/videos/{request_id}`	Asynchronous video generation (Grok Imagine), see Images & video	Per second, settled once the video is ready
`POST /v1/responses/compact` · `/v1/responses/input_tokens` · `/v1/alpha/search`	Context compaction, input-token estimation and web search used by the Codex client	Compaction per token; estimation free; search per call
`GET /v1/usage`	Account balance and total spend, see Billing & balance	Free, counts toward RPM

embeddings.sh

curl https://soleapi.com/v1/embeddings \
  -H "Authorization: Bearer $SOLEAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-3-small",
    "input": "The quick brown fox"
  }'

POST /v1/chat/completions is kept for legacy compatibility only. All new integrations should use the Responses API.