Duck.ai API

GET   /api/v1/search?engine=duck_ai

Send a question to the duck_ai engine to receive a structured answer. Text answers include Markdown and text blocks. Additional result collections depend on the source answer.

To continue an answer, pass pagination.conversation_token with the next question. Keep the same model for the conversation. Web search and reasoning availability depend on the selected model. Image-only answers do not return a conversation token.

API Parameters

Search Query

  • Name
    q
    Required
    Required
    Description

    The question to send to Duck.ai, up to 16000 characters. The limit counts UTF-16 code units; a supplementary character such as an emoji counts as two. For a long question, use a POST request with the parameters in the JSON body instead of the URL query string.

Answer Options

  • Name
    model
    Required
    Optional
    Description

    The model to use. The default parameter is gpt-6-luna. A follow-up keeps its conversation model. Supported values: gpt-6-luna, gpt-5.6-luna, gpt-5.4-mini, claude-haiku-5-5, claude-haiku-4-5, mistral-small-2603, tinfoil/gpt-oss-120b, tinfoil/gemma4-31b.

  • Name
    reasoning_effort
    Required
    Optional
    Description

    The reasoning effort supported by the selected model. gpt-6-luna: none, low; gpt-5.6-luna: none, low; gpt-5.4-mini: none, low; claude-haiku-5-5: none, low; claude-haiku-4-5: none, low; mistral-small-2603: omit reasoning_effort; tinfoil/gpt-oss-120b: low; tinfoil/gemma4-31b: none, low. When omitted, the first supported effort is used. Supported values: none, low.

Conversation

  • Name
    conversation_token
    Required
    Optional
    Description

    Use pagination.conversation_token from a fresh response to continue the conversation. Up to 64000 characters.

    Note: Tokens grow with the conversation (8KB+ after long answers). If you encounter 400, 413 or 414 errors, use a POST request with the token in the JSON body instead of the URL query string.

Engine

  • Name
    engine
    Required
    Required
    Description

    Set to duck_ai.

API key

  • Name
    api_key
    Required
    Required
    Description

    Your SearchAPI API key. You can also use the Authorization: Bearer header.

Zero Data Retention

  • Name
    zero_retention
    Enterprise Only
    Enterprise Only
    Required
    Optional
    Description

    Set to true to disable persistent storage and request logging.

Response

Response
GET
https://www.searchapi.io/api/v1/search?engine=duck_ai&model=gpt-6-luna&q=What+is+the+capital+of+France%3F
Request
import requests

url = "https://www.searchapi.io/api/v1/search"
params = {
  "engine": "duck_ai",
  "q": "What is the capital of France?",
  "model": "gpt-6-luna"
}

response = requests.get(url, params=params)
print(response.text)
Response
{
  "search_metadata": {
    "id": "search_ZD40AeqmorqVsJ65gkyw25lR",
    "status": "Success",
    "created_at": "2026-11-05T17:03:55Z",
    "request_time_taken": 3.95,
    "parsing_time_taken": 0.0,
    "total_time_taken": 3.96,
    "request_url": "https://duck.ai/",
    "html_url": "https://www.searchapi.io/api/v1/searches/search_ZD40AeqmorqVsJ65gkyw25lR.html",
    "json_url": "https://www.searchapi.io/api/v1/searches/search_ZD40AeqmorqVsJ65gkyw25lR"
  },
  "search_parameters": {
    "engine": "duck_ai",
    "q": "What is the capital of France?",
    "model": "gpt-6-luna"
  },
  "text_blocks": [
    {
      "type": "paragraph",
      "answer": "Paris."
    }
  ],
  "markdown": "Paris.",
  "response_metadata": {
    "model": "gpt-6-luna",
    "message_id": "msg_04d0b66b6df095c7016aecb6f4d65c87d1bff8bb32290bb1c0",
    "is_web_search_performed": false
  },
  "pagination": {
    "conversation_token": "eJxFi7EOgzAMBX_FejN07MDC1rlbB9TBAlMipQmyHakSyr8DSxnvTrfhmyeJ6PBZvb23sSRGAy-aDN2wQXOUoxYTPb38_KDXwk7ByBehkdfgHCnP9FBOo_Sozf9js2DOya_5yRrshvquOyvIKbc"
  }
}

Web Results

Web Results
GET
https://www.searchapi.io/api/v1/search?engine=duck_ai&model=gpt-5.6-luna&q=what+is+the+capital+of+japan&web_search=true
Request
import requests

url = "https://www.searchapi.io/api/v1/search"
params = {
  "engine": "duck_ai",
  "q": "what is the capital of japan",
  "model": "gpt-5.6-luna",
  "web_search": "true"
}

response = requests.get(url, params=params)
print(response.text)
Response
{
  "web_results": [
    {
      "position": 1,
      "title": "Japan",
      "link": "https://en.wikipedia.org/wiki/Japan",
      "source": "Wikipedia",
      "favicon": "https://external-content.duckduckgo.com/ip3/en.wikipedia.org.ico"
    },
    ...
  ]
}
Reference Links and Citations
GET
https://www.searchapi.io/api/v1/search?engine=duck_ai&model=gpt-6-luna&q=best+budget+laptops+for+students+2026
Request
import requests

url = "https://www.searchapi.io/api/v1/search"
params = {
  "engine": "duck_ai",
  "q": "best budget laptops for students 2026",
  "model": "gpt-6-luna"
}

response = requests.get(url, params=params)
print(response.text)
Response
{
  "reference_links": [
    {
      "index": 0,
      "title": "The best student laptops in 2026: our top tested picks for ...",
      "link": "https://www.techradar.com/computing/laptops/best-student-laptops",
      "source": "TechRadar",
      "date": "2026-09-29",
      "favicon": "https://external-content.duckduckgo.com/ip3/www.techradar.com.ico"
    },
    ...
  ],
  "citations": [
    {
      "index": 314,
      "source_indexes": [
        0
      ]
    },
    ...
  ]
}
Related Searches
GET
https://www.searchapi.io/api/v1/search?engine=duck_ai&model=gpt-5.4-mini&q=best+budget+laptops+for+students+2026
Request
import requests

url = "https://www.searchapi.io/api/v1/search"
params = {
  "engine": "duck_ai",
  "q": "best budget laptops for students 2026",
  "model": "gpt-5.4-mini"
}

response = requests.get(url, params=params)
print(response.text)
Response
{
  "related_searches": [
    {
      "query": "best student laptop 2026 budget",
      "link": "https://duckduckgo.com/?q=best%20student%20laptop%202026%20budget&t=duck_ai",
      "score": 0.95
    },
    ...
  ]
}

Inline Images

Inline Images
GET
https://www.searchapi.io/api/v1/search?engine=duck_ai&model=gpt-5.4-mini&q=generate+an+image+of+a+red+bicycle+on+a+beach
Request
import requests

url = "https://www.searchapi.io/api/v1/search"
params = {
  "engine": "duck_ai",
  "q": "generate an image of a red bicycle on a beach",
  "model": "gpt-5.4-mini"
}

response = requests.get(url, params=params)
print(response.text)
Response
{
  "inline_images": [
    {
      "title": "Drawing a red bicycle on a sunny beach",
      "original": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQAAAQABAAD/61wnSlACEQAAAAEAAFwdanVtYgAAAB5qdW1kYzJwYQARABCAA...",
      "width": 1536,
      "height": 1024,
      "query": "A realistic red bicycle standing on a sunny beach near the shoreline, with soft waves, golden sand, clear blue sky, and natural daylight. Composition centered o...",
      "model": "GPT Image 2"
    }
  ]
}