FreeAI::DuckAI - Duck.ai AI Service Scraper
Scraper overview
The Duck.ai scraper allows you to receive answers from leading AI models via the Duck.ai service by DuckDuckGo — for free and without registration. The service ensures complete anonymity: queries are not saved and are not used for model training.
The scraper supports natural language queries in any language, including English, Russian, and others.
Output results can be saved in any required format thanks to the powerful Template Toolkit engine, which allows structuring data into JSON, CSV, SQL, and other formats.
Collected data
- AI model response text
- Name of the model used
- Links, titles, sites, and publication dates of sources
Capabilities
- AI model selection: GPT-6 Luna, GPT-5.4 Mini, Claude Haiku 4.5, Mistral Small 4, GPT OSS 120B, Gemma 4 31B
- Collecting sources separately from the response text
- Proxy support
- Headless and headful browser modes
- Automatic retry on errors
Use cases
- Getting answers to any questions without registration and restrictions
- Generation of texts, descriptions, articles, and content
- Collecting structured answers for thematic queries to create knowledge bases
- Automating the processing of large lists of queries with AI answers
- Comparing answers from different AI models for the same queries
- Extracting links to sources with titles and publication dates
- Any other tasks where a quick AI response is required without registration
Queries
As queries, you should specify questions or tasks, exactly as if you were entering them directly into the Duck.ai field, for example:
What is A-Parser?
How to learn to learn fast?
How to improve memory and concentration?
Results
Here and below, result examples are shortened for better clarity
By default, the query and its answer are output:
$query
$answer
Example:
What is A-Parser?
A-Parser is a powerful tool for automatic data collection (scraping) from various sources on the Internet...
Sources are not included in this text. They are located in the $p1.sources array.
Result output options
Answer, model, and sources
Result format:
ANSWER:
$p1.answer
\n
MODEL:
$p1.model
\n
SOURCES:
$p1.sources.format('URL: $url \n Title: $title \n Site: $site \n Published: $published \n\n')
Example result:
ANSWER:
A-Parser is a powerful tool for automated data collection from various sources on the Internet...
MODEL:
gpt-6-luna
SOURCES:
URL:https://a-parser.com/
Title:A-Parser — professional data scraper
Site:a-parser.com
Published:
Exporting a list of links
Result format:
$p1.sources.format('$url\n')
Example result:
https://a-parser.com/
https://en.wikipedia.org/wiki/Web_scraping
CSV output of links, titles, sites, and dates
Result format:
[% FOREACH item IN p1.sources;
tools.CSVline(item.url, item.title, item.site, item.published);
END %]
Example result:
https://a-parser.com/,"A-Parser — professional data scraper",a-parser.com,
https://en.wikipedia.org/wiki/Web_scraping,"Web scraping",Wikipedia,"October 1, 2026"
Specify the csv extension in the result file name.
JSON output
Common result format:
[% IF notFirst;
",\n";
ELSE;
notFirst = 1;
END;
obj = {};
obj.query = query;
obj.answer = p1.answer;
obj.model = p1.model;
obj.json %]
Initial text:
[
Final text:
]
Result example:
[{"query":"What is A-Parser?","answer":"A-Parser is a powerful tool...","model":"gpt-6-luna"}]
Possible settings
| Parameter name | Default value | Description |
|---|---|---|
| Model | GPT-6 Luna (fast) | AI model selection: GPT-6 Luna, GPT-5.4 Mini, Claude Haiku 4.5, Mistral Small 4, GPT OSS 120B, Gemma 4 31B |
| Headless browser | ☑ | If the option is enabled, the browser will not be displayed during operation |
| Max concurrent pages | 10 | Maximum number of simultaneously open browser pages |