Skip to content

LongTrainer Logo

LongTrainer 1.0.0 — Production-Ready RAG Framework

Multi-tenant bots, streaming, tools, and persistent memory — all batteries included.

PyPI Version Total Downloads Monthly Downloads

Welcome to LongTrainer 1.0.0

LongTrainer is a production-ready RAG framework that turns your documents into intelligent, multi-tenant chatbots with minimal code. Built on top of LangChain, it handles multi-bot isolation, persistent MongoDB memory, FAISS vector search, streaming responses, custom tool calling, chat encryption, and vision support.

Quick Start

Install LongTrainer and start building in minutes:

pip install longtrainer

RAG Mode (Default)

from longtrainer.trainer import LongTrainer
import os

os.environ["OPENAI_API_KEY"] = "sk-..."

trainer = LongTrainer(mongo_endpoint="mongodb://localhost:27017/")
bot_id = trainer.initialize_bot_id()

trainer.add_document_from_path("data.pdf", bot_id)
trainer.create_bot(bot_id)

chat_id = trainer.new_chat(bot_id)
answer, sources = trainer.get_response("What is this about?", bot_id, chat_id)
print(answer)

Agent Mode (With Tools)

from longtrainer.tools import web_search
from langchain_core.tools import tool

@tool
def calculate(expression: str) -> str:
    """Evaluate a math expression."""
    return str(eval(expression))

trainer.add_tool(web_search, bot_id)
trainer.add_tool(calculate, bot_id)

trainer.create_bot(bot_id, agent_mode=True)
chat_id = trainer.new_chat(bot_id)
answer, _ = trainer.get_response("What is 42 * 17?", bot_id, chat_id)

What's New in 1.0.0

  • Dual Mode: RAG (LCEL) for simple Q&A, Agent (LangGraph) for tool calling
  • Streaming Responses: Sync and async out of the box
  • Custom Tool Calling: add_tool() with any LangChain @tool
  • Per-Bot Customization: Independent LLM, embeddings, and retrieval config per bot
  • Chat Encryption: Fernet encryption for stored conversations

Upgrading from 0.3.4? See the Migration Guide.

Documentation

Guide Description
Installation Install LongTrainer and system dependencies
Creating an Instance Configure the LongTrainer class
Creating and Using a Bot Bot lifecycle: create, load, chat
Agent Mode & Tools Tool calling, streaming, agent configuration
Chat Management Sessions, history, training on chats
Supported Formats Document types and ingestion methods
Integrating LLMs Use any LangChain-compatible LLM
Integrating Embeddings Custom embedding models
Updating Bots Add documents and reconfigure bots
Deleting Bots Remove bots and associated data
Migration 0.3.4 → 1.0.0 Breaking changes and upgrade path