Question 1

What is Modelup?

Accepted Answer

Modelup is a free, open-source, browser-only workspace for discovering, designing, and running local large language model (LLM) pipelines. It combines a model catalog, a node-based pipeline designer, a hardware profiler, and a chat console for local models in one installable web app.

Question 2

Is Modelup free and open source?

Accepted Answer

Yes. Modelup is free to use and its source code is publicly available on GitHub. It is built with React, Vite, TypeScript, and Bun.

Question 3

Does Modelup send my data to the cloud?

Accepted Answer

No. Modelup is local-first. There is no backend server, no user accounts, and no authentication. Conversations, pipelines, and settings are stored in your browser's localStorage. Inference happens on your own machine via Ollama, so prompts and model data never leave your device.

Question 4

What is a local LLM?

Accepted Answer

A local LLM is a large language model that runs entirely on your own hardware rather than a remote cloud API. This keeps your data private, removes per-token costs, and lets you work offline. Modelup helps you discover, design around, and chat with such models.

Question 5

Do I need a GPU to use Modelup?

Accepted Answer

No. Modelup itself runs in any modern browser and needs no GPU. Whether a specific model runs smoothly depends on your hardware and Ollama. The built-in hardware profiler reads your RAM, CPU cores, and GPU and suggests compatible model sizes.

Question 6

What is Ollama and how does Modelup use it?

Accepted Answer

Ollama is a tool that runs LLMs locally on your machine. Modelup connects to your local Ollama instance at localhost:11434 to list installed models, pull new ones, and stream chat responses. You install and control Ollama yourself; Modelup does not bundle or replace it.

Question 7

Can I design LLM pipelines in Modelup?

Accepted Answer

Yes. The pipeline designer is a node-based canvas where you compose stages such as Model, Prompt, Post-process, and Output. The graph is kept acyclic, can be saved to your browser, loaded from starter templates, and exported as JSON.

Question 8

How do I install and run Modelup locally?

Accepted Answer

Clone the repository, run bun install, then bun dev and open the printed URL. For local model chat, install Ollama and let it run on localhost:11434.