Extensions Overview
The current Jan Desktop Client has some default extensions built on this framework to enhance the user experience. This guide will show you the list of default extensions and how to configure extension settings.
Default Extensions
The default extensions are in the Settings
> Extensions
.
List of Default Extensions
Extension Name | Version | Description | Source Code Link |
---|---|---|---|
Assistant Extension | v1.0.0 | This extension enables assistants, including Jan, a default assistant that can call all downloaded models. | Source (opens in a new tab) |
Conversational Extension | v1.0.0 | This extension enables conversations and state persistence via your filesystem. | Source (opens in a new tab) |
Inference Engine Nitro Extension | v1.0.0 | This extension embeds Nitro, a lightweight (3 MB) inference engine in C++. See https://nitro.jan.ai (opens in a new tab). | Source (opens in a new tab) |
Inference Engine OpenAI Extension | v1.0.0 | This extension enables OpenAI chat completion API calls. | Source (opens in a new tab) |
Inference Engine Triton TRT-LLM Extension | v1.0.0 | This extension enables Nvidia's TensorRT-LLM as an inference engine option. | Source (opens in a new tab) |
Inference Engine Tensor RT Llm Extension | v0.0.3 | This extension enables Nvidia's TensorRT-LLM for the fastest GPU acceleration. | Source (opens in a new tab) |
Inference Engine MistralAI Extension | v1.0.0 | This extension enables Mistral chat completion API calls. | Source (opens in a new tab) |
Inference Engine Groq Extension | v1.0.0 | This extension enables Groq chat completion API calls. | Source (opens in a new tab) |
HuggingFace Extension | v1.0.0 | This extension converts HF models to GGUF. | Source (opens in a new tab) |
Model Management Extension | v1.0.30 | Model Management Extension provides model exploration and seamless downloads. | Source (opens in a new tab) |
System Monitoring Extension | v1.0.10 | This extension offers system health and OS-level data. | Source (opens in a new tab) |
Configure Extension Default Settings
To configure extension default settings:
- Navigate to the
~/jan/extensions
. - Open the
extensions.json
file - Edit the file with options including:
Option | Description |
---|---|
_active | Enable/disable the extension. |
listeners | Default listener setting. |
origin | Extension file path. |
installOptions | Version and metadata configuration. |
name | Extension name. |
productName | Extension display name. |
version | Extension version. |
main | Main file path. |
description | Extension description. |
url | Extension URL. |
{ "@janhq/conversational-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-conversational-extension-1.0.0.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/conversational-extension", "productName": "Conversational", "version": "1.0.0", "main": "dist/index.js", "description": "This extension enables conversations and state persistence via your filesystem", "url": "extension://@janhq/conversational-extension/dist/index.js" }, "@janhq/model-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-model-extension-1.0.30.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/model-extension", "productName": "Model Management", "version": "1.0.30", "main": "dist/index.js", "description": "Model Management Extension provides model exploration and seamless downloads", "url": "extension://@janhq/model-extension/dist/index.js" }, "@janhq/inference-mistral-extension": { "_active": false, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-inference-mistral-extension-1.0.0.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/inference-mistral-extension", "productName": "MistralAI Inference Engine", "version": "1.0.0", "main": "dist/index.js", "description": "This extension enables Mistral chat completion API calls", "url": "extension://@janhq/inference-mistral-extension/dist/index.js" }, "@janhq/inference-groq-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-inference-groq-extension-1.0.0.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/inference-groq-extension", "productName": "Groq Inference Engine", "version": "1.0.0", "main": "dist/index.js", "description": "This extension enables fast Groq chat completion API calls", "url": "extension://@janhq/inference-groq-extension/dist/index.js" }, "@janhq/inference-openai-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-inference-openai-extension-1.0.0.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/inference-openai-extension", "productName": "OpenAI Inference Engine", "version": "1.0.0", "main": "dist/index.js", "description": "This extension enables OpenAI chat completion API calls", "url": "extension://@janhq/inference-openai-extension/dist/index.js" }, "@janhq/inference-triton-trt-llm-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-inference-triton-trt-llm-extension-1.0.0.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/inference-triton-trt-llm-extension", "productName": "Triton-TRT-LLM Inference Engine", "version": "1.0.0", "main": "dist/index.js", "description": "This extension enables Nvidia's TensorRT-LLM as an inference engine option", "url": "extension://@janhq/inference-triton-trt-llm-extension/dist/index.js" }, "@janhq/huggingface-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-huggingface-extension-1.0.0.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/huggingface-extension", "productName": "HuggingFace", "version": "1.0.0", "main": "dist/index.js", "description": "Hugging Face extension for converting HF models to GGUF", "url": "extension://@janhq/huggingface-extension/dist/index.js" }, "@janhq/monitoring-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-monitoring-extension-1.0.10.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/monitoring-extension", "productName": "System Monitoring", "version": "1.0.10", "main": "dist/index.js", "description": "This extension provides system health and OS level data", "url": "extension://@janhq/monitoring-extension/dist/index.js" }, "@janhq/assistant-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-assistant-extension-1.0.1.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/assistant-extension", "productName": "Jan Assistant", "version": "1.0.1", "main": "dist/index.js", "description": "This extension enables assistants, including Jan, a default assistant that can call all downloaded models", "url": "extension://@janhq/assistant-extension/dist/index.js" }, "@janhq/tensorrt-llm-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-tensorrt-llm-extension-0.0.3.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/tensorrt-llm-extension", "productName": "TensorRT-LLM Inference Engine", "version": "0.0.3", "main": "dist/index.js", "description": "This extension enables Nvidia's TensorRT-LLM for the fastest GPU acceleration. See the [setup guide](https://jan.ai/guides/providers/tensorrt-llm/) for next steps.", "url": "extension://@janhq/tensorrt-llm-extension/dist/index.js" }, "@janhq/inference-nitro-extension": { "_active": true, "listeners": {}, "origin": "C:\\Users\\ACER\\AppData\\Local\\Programs\\jan\\resources\\app.asar.unpacked\\pre-install\\janhq-inference-nitro-extension-1.0.0.tgz", "installOptions": { "version": false, "fullMetadata": true }, "name": "@janhq/inference-nitro-extension", "productName": "Nitro Inference Engine", "version": "1.0.0", "main": "dist/index.js", "description": "This extension embeds Nitro, a lightweight (3mb) inference engine written in C++. See https://nitro.jan.ai.\nAdditional dependencies could be installed to run without Cuda Toolkit installation.", "url": "extension://@janhq/inference-nitro-extension/dist/index.js" }}
Specific Extension Settings
Jan offers an Extensions settings menu for configuring extensions that have registered their settings within the application. Here, you can directly integrate Remote Inference Engines with Jan without inserting the URL and API Key directly in the JSON
file. Additionally, you can turn the Logging extensions available on or off in Jan. To access the Extension settings, follow the steps below:
- Navigate to the main dashboard.
- Click the gear icon (⚙️) on the bottom left of your screen.
- The registered extension settings can be customized under the Extensions section.
System Monitor Extension Feature
The System Monitor extension now offers enhanced customization for app logging. Users can toggle the application logging feature on or off and set a custom interval for clearing the app logs. To configure the app log feature, follow these steps:
- Navigate to the main dashboard.
- Click the Gear Icon (⚙️) on the bottom left of your screen.
- Under the Extensions section, select the System Monitoring extension.
- Use the slider to turn the app logging feature on or off.
- Specify the log cleaning interval in milliseconds.
- You can clear the app logs manually by clicking the Clear logs button in the advanced settings.
- There are no minimum or maximum intervals for setting the time. However, invalid inputs will default to
120000ms (2 minutes)
.