Firellama par dishdare

Right-click to summarize a sentence or define a word using locally installed ollama. Sends selected text to your offline ollama system for pop-up window. This is more or less DIY project since most of the things are needed on your local machine.

0 (0 reviews)

1 User

Vous avez besoin de Firefox pour utiliser cette extension

Télécharger Firefox et obtenir l’extension

Télécharger le fichier

À propos de cette extension

you need ollama installed on your local machine.
https://ollama.com/download
ollama pull deepseek-r1:1.5b

The addon will use the model deepseek-r1:1.5b installed on your machine.
It summarizes the selected text if it is more than one word or else defines it.

On top of that you need server.py running on your machine.
pip install flask flask_cors

server.py

from flask import Flask, request, jsonify
from flask_cors import CORS
import requests
import re

app = Flask(__name__)
CORS(app)

OLLAMA_URL = "http://localhost:11434/api/generate"
OLLAMA_MODEL = "deepseek-r1:1.5b"

def call_ollama(prompt):
    try:
        res = requests.post(OLLAMA_URL, json={
            "model": OLLAMA_MODEL,
            "prompt": prompt,
            "stream": False
        })
        if res.status_code == 200:
            result = res.json().get("response", "")
            cleaned = re.sub(r"<think>.*?</think>", "", result, flags=re.DOTALL).strip()
            return cleaned
        else:
            return f"⚠️ Ollama error: {res.text}"
    except Exception as e:
        return f"❌ Error: {str(e)}"

@app.route("/ask", methods=["POST"])
def ask():
    data = request.get_json()
    prompt = data.get("prompt")
    if not prompt:
        return jsonify({"error": "Missing prompt"}), 400

    result = call_ollama(prompt)
    return jsonify({"text": result})

if __name__ == "__main__":
    app.run(host="0.0.0.0", port=5005)

Run it at startup with "ollama run deepseek-r1:1.5b" and enjoy the summary.

I usually do ctrl+a of the entire webpage and right-click on Firellama to get the summary.

Special Thanks to ChatGPT.

Noté 0 par 1 personne