Just loading, one second!

Make your own AI Chatbot

⦁ 3 minute read.


AI chatbots like ChatGPT are pretty cool, no one’s denying that, but also they train it on your inputs by default and you’re at the mercy of OpenAI/Google/Other AI Overlord as to when you can use it. Why not make/host your own?


If you haven’t got Python installed, you probably want to install miniconda.

We need to install a few things first (scope it to the project):

and you also want to install a llama model. This is more complicated than it needs to be, but you sign up to Facebook’s Meta’s llama access here and then download one of the .gguf files here. The one I’m using is called llama-2-13b-chat.Q4_K_M.gguf.


Chainlit is super easy to use, if you want an interactive app with most of the work done for you, then that’s enough!

If you create a file called app.py, pop this into it

import chainlit as cl

async def main(message: str):
    await cl.Message(content=f"Received: {message}").send()

and then run the following command

chainlit run app.py -w

The -w flag turns on auto-reloading, so you can edit app.py and it’ll update live at http://localhost:8000 (not https for some reason).


from langchain.llms import LlamaCpp

llm = LlamaCpp(
    model_path="./models/llama-2-13b-chat.Q4_K_M.gguf", # Make sure you change this to the path you downloaded it to earlier

output = llm("Q: Name the planets in the solar system? A: ", max_tokens=32, stop=["Q:", "\n"], echo=True)


  "id": "cmpl-xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx",
  "object": "text_completion",
  "created": 1679561337,
  "model": "./models/7B/ggml-model.bin",
  "choices": [
      "text": "Q: Name the planets in the solar system? A: Mercury, Venus, Earth, Mars, Jupiter, Saturn, Uranus, Neptune and Pluto.",
      "index": 0,
      "logprobs": None,
      "finish_reason": "stop"
  "usage": {
    "prompt_tokens": 14,
    "completion_tokens": 28,
    "total_tokens": 42

Combining them

Now if we stick them together, we find this, a pretty simple minimal working example for a chat bot.

import chainlit as cl
from langchain.llms import LlamaCpp
from langchain import PromptTemplate, LLMChain

template = """You are a helpful and friendly AI chatbot. Try and answer the following if you can.
Question: {question}
Answer: """

def main():
    # Make sure the model path is correct for your system!
    llm = LlamaCpp(

    llm_chain = LLMChain(prompt=prompt, llm=llm, verbose=True)

    # Store the chain in the user session
    cl.user_session.set("llm_chain", llm_chain)

async def main(message: str):
    # Get chain from the user session
    llm_chain = cl.user_session.get("llm_chain")

    # Call the chain asynchronously
    res = await llm_chain.acall(message)

    await cl.Message(content=res["text"]).send()