Deploy your Nuxt project to Vercel with zero configuration.
Live Example: https://nuxtjs-template.vercel.app
Look at the Nuxt 3 documentation to learn more.
-
have a locally running instance of ollama
-
or have it runnning in colab with the following link: https://colab.research.google.com/drive/1oSSNt6Osme6FRXpRy9wTRKi2t433EQXB?usp=sharing
-
have pulled the
dolphin-mistral:7b-v2.6-dpo-laser-q5_1on that instance (or pull through the ui later)
Then you can either
- start the server as written below or
- use the app directly via https://hatespeech-detector.vercel.app/
If you get a connection error, this is probably because you didn't connect the ollama instance properly
- check if the ollama instance is running
- check the console of the browser for detailed errors
- make sure the correct model is pulled on the ollama instance (and finished. The model is a few GB in size)
- pulling through the ui seems to sometimes have issues, ideally pull the model directly with
ollama pull [modelname] - you might have to run
ollama serveto expose the api
Make sure to install the dependencies:
# yarn
yarn
# npm
npm install
# pnpm
pnpm install --shamefully-hoistStart the development server on http://localhost:3000
npm run devBuild the application for production:
npm run buildLocally preview production build:
npm run previewCheckout the deployment documentation for more information.