An open API service indexing awesome lists of open source software.

https://github.com/posit-dev/chatlas


https://github.com/posit-dev/chatlas

chatbot llm python

Last synced: 3 months ago
JSON representation

Awesome Lists containing this project

README

          

# chatlas chatlas website

PyPI
MIT License
versions
Python Tests

Your friendly guide to building LLM chat apps in Python with less effort and more clarity.

## Install

Install the latest stable release [from PyPI](https://pypi.org/project/chatlas/):

```bash
pip install -U chatlas
```

Or, install the latest development version from GitHub:

```bash
pip install -U git+https://github.com/posit-dev/chatlas
```

## Quick start

Get started in 3 simple steps:

1. Choose a model provider, such as [ChatOpenAI](https://posit-dev.github.io/chatlas/reference/ChatOpenAI.html) or [ChatAnthropic](https://posit-dev.github.io/chatlas/reference/ChatAnthropic.html).
2. Visit the provider's [reference](https://posit-dev.github.io/chatlas/reference) page to get setup with necessary credentials.
3. Create the relevant `Chat` client and start chatting!

```python
from chatlas import ChatOpenAI

# Optional (but recommended) model and system_prompt
chat = ChatOpenAI(
model="gpt-4.1-mini",
system_prompt="You are a helpful assistant.",
)

# Optional tool registration
def get_current_weather(lat: float, lng: float):
"Get the current weather for a given location."
return "sunny"

chat.register_tool(get_current_weather)

# Send user prompt to the model for a response.
chat.chat("How's the weather in San Francisco?")
```

Model response output to the user query: 'How's the weather in San Francisco?'

Learn more at