🤖Cllama

Cllama is a python streamlit webUI for using LLM models which are available in the ollama localy.

Features

• Real Time weather information using tools.
• Real Time weather alert using tools.

Installation Guide

1. Install Ollama locally:

   you can install using official doc https://ollama.com/download or you can run the install_ollama.sh bash file.Follow these commands:

• chmod +x install_ollama_models.sh

• ./install_ollama_models.sh

2. Download a LLm model on Ollama

   Download any llm model's from the Ollama model Library https://ollama.com/search. for instance, to install IBM's Granite 3.3 model, use this command:

• ollama run granite3.3

To see the downloaded or available LLM models in the local Ollama, use this command:

• ollama list

3. Install UV dependency manager for python.

   use this command to install UV using python's pip

• pip install uv

use this command to install using Home Brew (recommended for Mac OS).

• brew install uv

4. Install cllama dependencies using UV.

To sync all the dependencies for the cllama project, use

• uv sync

5. Runing cllama

To run the cllama webUI project use this streamlit command from the base directory.

• streamlit run src/main.py

Contribution Guidelines

a) Create a task in the project planning board (CPPB) https://github.com/users/joshanjohn/projects/17
b) use proper branch name as github task suggest, which is hash number followed by task name seperated by hypen. (eg: 4-task-title)
c) Submit a pull request (PR) before merging into main branch.
d) Must get an PR approval review to merge.

Lint

1. In your terminal, cd into the project root
2. Run: uv run ruff check
3. To apply safe automatic fix, run: uv run ruff check --fix.
4. To apply unsafe automatic fix, run: uv run ruff check --fix --unsafe-fixes

You may need to run formatter black . again after running ruff auto fix.

Format

1. In your terminal, cd into the project root
2. Run: uv run black .

All The Best 🙌🏻