So-Sahari / jenn-ai

Generative AI tool with Go + HTMX

JennAI

JennAI is an open-source chat interface to interact with LLMs. Part of the inspiration of the project was getting a Go generative AI project running. After getting a small Go + HTMX + Ollama project setup, I wanted to add more features to it. JennAI is the generative AI helper that gives both cli and UI access to LLMs.

Features

Support for models is currently limited to:

• Bedrock Foundational Models
• Ollama Models

Dependencies

• Bedrock requires model access set as default profile or leveraging AWS_PROFILE
• Ollama requires installation link

Installation

git clone <repo>
go install ./cmd/jenn-ai

Local Development

If you have go installed, ensure that you have air

make deps

Run in development to perform live edits

make local_dev

Run locally

make local

Docker

Run locally (in development)

make build_dev
make up_dev

Run locally

make build
make up

To bring it down, run

make down

If you have a GPU, then you need to modify the input

make up COMPOSE_FILE=docker-compose-gpu.yaml

Make sure to include the variable when running down