# AI use cases for Jan (janhq/jan)

Repo: [janhq/jan](https://github.com/janhq/jan) | Use cases: 3 | Page: https://dedupedai.com/use-cases/janhq-jan/

Jan is an open-source ChatGPT replacement that downloads and runs LLMs such as Llama, Gemma, Qwen and GPT-oss locally from Hugging Face, with full control and privacy. It can also connect to cloud models from OpenAI, Anthropic, Mistral, Groq and others, lets you create custom assistants, exposes an OpenAI-compatible API at localhost:1337 and supports the Model Context Protocol for agentic features.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Chat with an AI model fully offline: https://dedupedai.com/use-cases/janhq-jan/chat-with-a-local-model-offline/
- Use Jan as a local OpenAI-compatible API: https://dedupedai.com/use-cases/janhq-jan/use-jan-as-a-local-openai-api/
- Build Jan from source with your GPU engine: https://dedupedai.com/use-cases/janhq-jan/build-jan-from-source-with-gpu-support/

## How to chat with an AI model fully offline with Jan

Repo: [janhq/jan](https://github.com/janhq/jan) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README

Install Jan, download an open model from Hugging Face and chat without sending anything to the cloud.

**Who it is for:** Someone who wants ChatGPT-style help with private notes or documents without a subscription.

**The problem:** Cloud chatbots send every message to a third party, and setting up local models usually takes a terminal.

**The outcome:** A desktop chat app running an open model locally, with everything kept on your machine.

### Steps

1. Download Jan for your OS from Jan.ai or GitHub Releases: Jan.exe, Jan.dmg, Jan.deb or Jan.AppImage
2. Open Jan and download a model such as Llama, Gemma or Qwen from Hugging Face
3. Pick a model size your RAM can handle, for example 3B on an 8GB Mac
4. Start chatting; everything runs locally when you choose a local model

### Prompt to paste into your agent

```text
Help me set up janhq/jan on my <OS> computer with <RAM> of memory. Recommend a local model size that will run smoothly and walk me through downloading it and starting a chat.
```

### Caveat

Speed and quality depend on your hardware; larger models need much more RAM.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/janhq-jan/chat-with-a-local-model-offline/

## How to use Jan as a local OpenAI-compatible API with Jan

Repo: [janhq/jan](https://github.com/janhq/jan) | Difficulty: Beginner | Time: 20 minutes | Basis: From the README

Turn on Jan's local server and point your scripts or apps at localhost:1337.

**Who it is for:** A developer who wants to test an app against a local model without paying per token.

**The problem:** Building against a cloud API means costs and sending data out during development.

**The outcome:** A local endpoint at localhost:1337 that apps written for the OpenAI API can call.

### Steps

1. Install Jan and download a local model
2. Enable the OpenAI-Compatible API so the local server runs at localhost:1337
3. Point your app's OpenAI base URL at the local server
4. Check the API Reference linked in the README for endpoints

### Prompt to paste into your agent

```text
Using janhq/jan's OpenAI-compatible server at localhost:1337, change my <script or app> to call the local model <model> instead of the OpenAI API, run it, and compare the answer to the cloud version.
```

### Caveat

Local models are usually weaker than frontier cloud models, so expect differences in quality.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/janhq-jan/use-jan-as-a-local-openai-api/

## How to build Jan from source with your GPU engine with Jan

Repo: [janhq/jan](https://github.com/janhq/jan) | Difficulty: Advanced | Time: 1 hour | Basis: From the README

Clone Jan and build it with make, choosing a CUDA, Vulkan, Metal or ROCm engine variant.

**Who it is for:** A developer who wants to contribute to Jan or needs a build tuned for their GPU.

**The problem:** Prebuilt binaries may not match your hardware or let you change the code.

**The outcome:** A development build of Jan running locally with the engine variant you chose.

### Steps

1. Install Node.js 20+, Yarn 4.5.3+, Make and Rust
2. Clone with `git clone https://github.com/janhq/jan` and `cd jan`
3. Run `make dev` to install dependencies, build and launch the app
4. Pick an engine variant, for example `make dev JAN_ENGINE_VARIANT=cuda13`
5. On Windows, run make from Git Bash, not cmd.exe

### Prompt to paste into your agent

```text
Build janhq/jan from source on this machine with the <engine variant> engine using make dev. If the build fails, read the error and apply the fix the README gives for that case.
```

### Caveat

On Windows, CUDA builds can hit the 260 character path limit; the build script relocates the engine build tree, and otherwise the README says to set JAN_ENGINE_BUILD_DIR to a short path.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/janhq-jan/build-jan-from-source-with-gpu-support/