ai-ollama-py

Project Description

ai-ollama-py is a Flask server built with Python 3.10 that processes JSON payloads containing context and prompt strings. The server interacts with Ollama on port 11434 and exposes three endpoints:

/api/admin_engine
/api/arc_engine
/api/api_engine

Sample JSON Payload

The server expects a JSON payload in the following format:

  {
    "context": "your context here",
    "prompt": "your prompt here"
  }

Prerequisites

Python 3.10: Ensure you have Python 3.10 installed. You can download it from the official Python website.
Ollama: Ollama is a dependency for this project. It's recommended to run Ollama on a machine with a GPU for better performance, but there is also a Dockerized option.

Setup

Ollama Installation

Download Ollama:

curl -fsSL https://ollama.com/install.sh | sh

Run Ollama: After installation, start the Ollama server by running the following command in your terminal:
```
ollama serve
```
Pull Llama3:
```
ollama pull llama3
```
*Pulling llama3 for the first time may take a few minutes

**Please note that ollama runs best on a machine with a dedicated GPU, you may run this on a CPU only machine but expect slower response times

Install Python Requirements

Use python 3.10

Run

python -m pip install -r requirements.txt

Run The Server

  python main.py

query example

curl -X POST -H "Content-Type: application/json" -d '{"context": "block ui","prompt": "How can i create a volume using the ui"}' http://0.0.0.0:8087/api/admin_engine

Contact

Tomer Gafsou

LOCAL TO BE REMOVED

export INGEST_DIR=/home/danduh/dev/mist-portal-nrwl/apps/portal-e2e/
python3 main.py
----
OLLAMA_HOST=0.0.0.0 ollama serve

{
  "model": "gpt-4-turbo",
  "temperature": 0.1,
  "messages": [
    {
      "role": "system",
      "content": ""
    },
    {
      "role": "user",
      "content": "Create Page Object class with name . \n INPUT>>:  <<INPUT  \n undefined"
    }
  ]
}

danduh / ai-ollama-py