SADIK AI

SADIK AI

A multilingual chat assistant that runs fully inside your browser through WebAssembly, so your messages stay on your device. Also reachable through Ollama and a local API.

Made by SadikNawazAI on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: SADIK AI is a multilingual chat assistant by Sadik Nawaz. This Space runs the whole model inside your web browser, using the wllama WebAssembly engine built on llama.cpp. No server does the thinking, and the README says it works on phones and on desktops.

The model behind it is named SadikNawazAI/sadik-ai-model. The README also shows other ways to call the same model from code.

Who it suits: It suits people who want a chat assistant that stays on their own device, and developers who want to see a language model run in a web page.

Industry | Who uses it | How they use it | Example
Everyday use | Multilingual users | Chat with an assistant in their own language | A user greets the assistant in Hindi and reads the reply
Software | Web developers | Study how a model runs through WebAssembly in a page | A developer reads app.js and the wllama folder
Education | Students | Try a language model without any server or account | A student opens the page on a laptop and asks a question
AI engineering | Developers | Call the model through Ollama or a local chat API | A developer runs it with one Ollama command

You can skip it if you need a hosted API, since the README's API examples point to servers you run yourself.

Where to use it: You use it in these places.

• A web browser on a computer or a phone, on the Space's page on Hugging Face
• A local copy of the page, opened in a browser on Windows, macOS or Linux
• Ollama, using the model's Hugging Face name
• A local server that offers a chat completions endpoint
• Google Colab, which the README offers as a free GPU API server

Use case | Who | What they get back
Chat in the browser | Anyone | Replies from the model with no server involved
Call it from code | Developers | Chat replies through a local Ollama or chat completions endpoint
Borrow a GPU | Developers | A Colab notebook from the model repository that serves an API

How to set it up: Online use needs only a browser. A local copy needs Python for the download step.

• A web browser, on a computer or a phone
• Python with pip, only to fetch a local copy
• Ollama, only if you want the one-command route

1. Open the Space's page on Hugging Face to use it online, with nothing to install.
2. For a local copy, install the Hugging Face Hub package.
3. Download the Space's files into a folder and move into it.
4. Load index.html in your browser.

pip install -U huggingface_hub
hf download SadikNawazAI/Sadik-AI-Chat --repo-type space --local-dir Sadik-AI-Chat
cd Sadik-AI-Chat

How to use it: Open the page and type a message in the chat box. The model answers in the same page. For the Ollama route, the README gives this single command.

ollama run hf.co/SadikNawazAI/sadik-ai-model

The README also shows a curl call to a local Ollama chat endpoint, and a Python call to a local chat completions endpoint that uses the model name sadik-ai. Both point to your own machine.

Example: You open the page and type a short greeting, such as asking who the assistant is. The model replies in the chat box, and the README says the exchange stays on your machine.

For newcomers and experts: A chat assistant normally runs on a company's server. This one runs inside your browser tab, so you can try it by just opening the page. Start with a short question in any language you like.

Under the hood, the engine is wllama, a WebAssembly build of llama.cpp, shipped in the repository's wllama folder, with the chat logic in app.js. It uses your CPU or GPU directly. The source does not say how big the model file is or where the browser fetches it from.

You open the page in a browser
wllama loads the model in WebAssembly
You type a message
The model answers on your own device

Good to know: The README makes several claims about privacy and cost, and these are its own words.

• Privacy: the README says no data leaves your machine.
• License: Apache 2.0, as the README states.
• Cost to run: none named. The README says zero server cost.
• Data it reaches: your messages, which the README says stay in the browser. The record names the model SadikNawazAI/sadik-ai-model.
• Not for: a hosted API, since the API examples target servers on your own computer.

On Hugging Face, 13 November 2026

• Likes: 2
• Built with: static
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/SadikNawazAI/Sadik-AI-Chat
• Files: https://huggingface.co/spaces/SadikNawazAI/Sadik-AI-Chat/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.