Overview

chatlas

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

Quick start

Get started in 3 simple steps:

Choose a model provider, such as ChatOpenAI or ChatAnthropic.
Visit the provider’s reference page to get setup with necessary credentials.
Create the relevant Chat client and start chatting!

from chatlas import ChatOpenAI

# Optional (but recommended) model and system_prompt
chat = ChatOpenAI(
    model="gpt-4o-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?")

# 🛠️ tool request
get_current_weather(37.7749, -122.4194)

# ✅ tool result
sunny

The current weather in San Francisco is sunny.

Install

Install the latest stable release from PyPI:

pip install -U chatlas

Why chatlas?

🚀 Opinionated design: most problems just need the right model, system prompt, and tool calls. Spend more time mastering the fundamentals and less time navigating needless complexity.

🧩 Model agnostic: try different models with minimal code changes.

🌊 Stream output: automatically in notebooks, at the console, and your favorite IDE. You can also stream responses into bespoke applications (e.g., chatbots).

🛠️ Tool calling: give the LLM “agentic” capabilities by simply writing Python function(s).

🔄 Multi-turn chat: history is retained by default, making the common case easy.

🖼️ Multi-modal input: submit input like images, pdfs, and more.

📂 Structured output: easily extract structure from unstructured input.

⏱️ Async: supports async operations for efficiency and scale.

✏️ Autocomplete: easily discover and use provider-specific parameters like temperature, max_tokens, and more.

🔍 Inspectable: tools for debugging and monitoring in production.

🔌 Extensible: add new model providers, content types, and more.

Next steps

Next we’ll learn more about what model providers are available and how to approach picking a particular model. If you already have a model in mind, or just want to see what chatlas can do, skip ahead to hello chat.