DuckDuckAI
DuckDuckAI is a Python package for interacting with DuckDuckGo's chat API. It allows you to fetch responses from DuckDuckGo's AI models and print them in a streamed format or as a complete response.
Installation
To install the DuckDuckAI package, you can use pip:
pip install DuckDuckAI
Usage
You can interact with DuckDuckAI by calling the ask function. It supports both streaming responses or returning the entire message at once.
Example
from duckduckai import ask
# Fetch response in streamed format (printing character by character)
ask("Tell me a joke", stream=True)[0] # Use 0 if you want to retrieve it inside a variable.
# Fetch response as a complete message
response = ask("Tell me a joke", stream=False)
print(response)
Parameters
| Parameter | Type | Description | Default |
|---|---|---|---|
| query | str | The search query string. | Required |
| stream | bool | Whether to stream results or fetch them all at once. | True |
| model | str | The model to use for the response. | gpt-4o-mini |
Available Models
DuckDuckAI currently supports the following models:
| Model ID | Description |
|---|---|
gpt-4o-mini |
A smaller variant of GPT-4o designed for quick, concise responses with less computation. |
meta-llama/Llama-3.3-70B-Instruct-Turbo |
Meta's large-scale Llama 3.3 model with 70 billion parameters designed for fast and accurate responses. |
claude-3-haiku-20240307 |
Anthropic's Claude 3 Haiku model optimized for efficient, high-quality responses. |
mistralai/Mistral-Small-24B-Instruct-2501 |
Mistral AI's 24 billion parameter model trained for instruction-based tasks. |
o3-mini |
OpenAI's compact reasoning model optimized for lightweight performance. |
Additional models may be available but subject to access restrictions. Some models may require specific permissions or may not be available in all regions.
Advanced Usage
You can reuse the authentication token to make multiple requests more efficiently:
from duckduckai import ask, fetch_x_vqd_token
# Fetch a token once
token = fetch_x_vqd_token()
# Use the same token for multiple requests
response1 = ask("What is quantum computing?", model="gpt-4o-mini", token=token)[0] # Do not put [0] if you want the token in the response
response2 = ask("Explain neural networks", model="claude-3-haiku-20240307", token=token)[0] # Do not put [0] if you want the token in the response
License
This project is licensed under the Apache-2.0 license - see the LICENSE file for details.
Metadata
Release files for DuckDuckAI 1.0.4
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| duckduckai-1.0.4.tar.gz | 3.8 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| DuckDuckAI-1.0.4-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 7.9 kB
Release files / duckduckai-1.0.4.tar.gz
| Download URL | duckduckai-1.0.4.tar.gz |
|---|---|
| Size | 3.8 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
f2820386fef4147be94abd984ca229eb16951e113428cc5e6bead09b35c5b4df
|
|
BLAKE2b-256 checksum How to use checksums |
d7712466e0fe302ce5c4e6e41f9eda6cc6bb2aa74a5d7e42195973721692fc0f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.5
|
Release files / DuckDuckAI-1.0.4-py3-none-any.whl
| Download URL | DuckDuckAI-1.0.4-py3-none-any.whl |
|---|---|
| Size | 4.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
050f10b57b41a45b4bfa3ada0c7e478a895f40139254f4165a6d3b2e401f6d89
|
|
BLAKE2b-256 checksum How to use checksums |
a039bac417b11a5a47827856117dfeba17ec2750481dd67a40d16f27982c57e9
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.5
|