# Model Management (/desktop/manage-models)
This guide shows you how to add, customize, and delete models within Jan.
## Local Model [#local-model]
Local models are managed through [Llama.cpp](https://github.com/ggerganov/llama.cpp), and these models are in a
format called GGUF. When you run them locally, they will use your computer's memory (RAM) and processing power, so
please make sure that you download models that match the hardware specifications for your operating system:
- [Mac](/desktop/install/mac#compatibility)
- [Windows](/desktop/install/windows#compatibility)
- [Linux](/desktop/install/linux#compatibility).
### Adding Models [#adding-models]
#### 1. Download from Jan Hub (Recommended) [#1-download-from-jan-hub-recommended]
The easiest way to get started is using Jan's built-in model hub (which is connected to [HuggingFace's Model Hub](https://huggingface.co/models):
1. Go to the **Hub** tab
2. Browse available models and click on any model to see details about it
3. Choose a model that fits your needs & hardware specifications
4. Click **Download** on your chosen model
Each model shows a colored fit pill — **Fits**, **May be slow**, or **Won't fit** — based on your hardware. Quantization tiers are grouped as **Small**, **Balanced**, or **Large**, with a **Recommended** tag on the default download. No data is downloaded to determine fit status.
Once you click **Download**, you can track the progress from the download indicator in the top bar. The model name and download percentage are shown in real time — you can cancel at any time by clicking the **×** button. If a download fails, it is cleared from the queue and a toast is shown. To retry, click **Download** again from the Hub or provider list.
#### 2. Import from Hugging Face [#2-import-from-hugging-face]
You can download models with a direct link from [Hugging Face](https://huggingface.co/):
**Note:** Some models require a Hugging Face Access Token. Enter your token in **Settings > General > Others > HuggingFace Token** before importing.
1. Visit the [Hugging Face Models](https://huggingface.co/models) page.
2. Find the model you want to use, make sure it is a GGUF file that fits in your computer.
3. Copy the **model ID** (e.g., TheBloke/Mistral-7B-v0.1-GGUF)
4. In Jan, paste the model ID to the **Search** bar in **Hub** page
5. Select your preferred quantized version to download (if the option is available)
**Copy the model ID.**
**Paste it in Jan's Hub Search Bar.**
#### 3. Use This Model on Hugging Face [#3-use-this-model-on-hugging-face]
You can also open any model directly in Jan from the Hugging Face model page:
1. On any model page, click **Use this model**
2. Under **Local Apps**, select **Jan**
#### Bulk Delete Downloaded Models [#bulk-delete-downloaded-models]
To free disk space quickly, open **Settings > Llama.cpp** (or **MLX**) and click **Delete All** in the Models section. Jan shows the total space to be freed and removes every managed download at once. Models you imported yourself are left untouched.
#### 4. Import Local Files [#4-import-local-files]
If you already have model files on your computer, you can import them directly via the engine settings:
}>Import GGUF model files for local inference on CPU or GPU.
}>Import MLX-Swift models for Apple Silicon.
## Cloud Models [#cloud-models]
Jan supports connecting to various AI cloud providers. Go to **Settings > Model Providers**, select your provider, and enter your API key.
}>Use GPT-4o, o3, and other OpenAI models.
}>Use Claude models from Anthropic.
}>Use Gemini models from Google.
}>Fast inference for open-source models.
}>Use Mistral AI models.
}>Access hundreds of models via a single API.