https://github.com/naskio/mergeui
All-in-one UI for merged LLMs in Hugging Face
https://github.com/naskio/mergeui
ai generative-ai huggingface large-language-models leaderboard llm merge mergekit mergoo model-merging
Last synced: 6 months ago
JSON representation
All-in-one UI for merged LLMs in Hugging Face
- Host: GitHub
- URL: https://github.com/naskio/mergeui
- Owner: naskio
- License: apache-2.0
- Created: 2024-04-23T12:21:01.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-06-10T06:48:11.000Z (over 1 year ago)
- Last Synced: 2025-04-03T09:50:59.943Z (6 months ago)
- Topics: ai, generative-ai, huggingface, large-language-models, leaderboard, llm, merge, mergekit, mergoo, model-merging
- Language: Python
- Homepage: https://naskio-mergeui.hf.space
- Size: 422 KB
- Stars: 24
- Watchers: 2
- Forks: 3
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
---
title: MergeUI
emoji: 🧬
colorFrom: purple
colorTo: pink
sdk: gradio
sdk_version: 4.31.5
python_version: 3.9
app_file: mergeui/web/gradio_app/main.py
fullWidth: true
header: mini
disable_embedding: false
startup_duration_timeout: 15m
license: apache-2.0
language:
- en
datasets:
- open-llm-leaderboard/results
tags:
- merge
- leaderboard
- mergekit
- lazymergekit
short_description: All-in-one UI for merged LLMs in Hugging Face Hub
thumbnail: https://raw.githubusercontent.com/naskio/mergeui/main/static/brand/banner.svg
pinned: true
---
[](https://naskio-mergeui.hf.space)
All-in-one UI for merged LLMs in Hugging Face
[MergeUI](https://naskio-mergeui.hf.space) is an [open-source project](https://github.com/naskio/mergeui) that aims to
provide a user-friendly interface for discovering and analyzing merged large language models (Merged LLMs) from
the [Hugging Face Hub](https://huggingface.co/models?other=merge&sort=trending).
It is particularly focused on models that have been merged using [MergeKit](https://github.com/arcee-ai/mergekit).
- 🚀 **[Try it out](https://naskio-mergeui.hf.space)**
- 👨💻 **[GitHub Repository](https://github.com/naskio/mergeui)**
- 🤗 **[Hugging Face Space](https://huggingface.co/spaces/naskio/mergeui)**
# Development
## Requirements
To run the project locally, you need to have the following installed:
- [Python 3.9+](https://www.python.org/)
- [Poetry 1.8+](https://python-poetry.org/)
- [Graphviz](https://www.graphviz.org/download/#mac)
- [Docker](https://www.docker.com/) with [Docker Compose](https://docs.docker.com/compose/)
## Setup
Once you have the requirements installed, you can set up the project by running the following commands:
```shell
poetry install
```
Once the dependencies are installed, make sure to set up the environment variables by creating a `.env` file
```shell
cp .env.example .env
```
> [!TIP]
> Find more about the environment variables in the [settings.py](./mergeui/core/settings.py) file.
Next, we need to launch the database and Redis using Docker Compose:
```shell
docker compose up -d
```
> [!TIP]
> run `docker compose down` to stop these services when you are done.
## Run
> [!NOTE]
> This project use `poetry` with `Poe the Poet` plugin to run commands, run `poe` to see all available commands.
Once the setup is complete, we need to index the models from the Hugging Face Hub and store them in the database.
### Indexing
For indexing models we use [RQ](https://python-rq.org/):
- First we need to launch some workers by running the following command in separate terminal tabs:
```shell
poe worker
```
- Next, we can start the indexing process by running:
```shell
poe index
```
- To monitor the indexing process, we can use the RQ dashboard by running:
```shell
rq-dashboard
```
> [!IMPORTANT]
> The indexing process takes few minutes to complete depending on your resources, number of workers and number
> of `merge` models available.
> [!NOTE]
> It takes around 6 minutes to index a graph of ~12k models and ~51k relationships using 64 workers.
### Visualisation
Once the indexing process is complete, we can start our app using the following command:
#### FastAPI server with a Gradio app
```shell
uvicorn mergeui.main:app --port 8000 --log-level trace
```
#### FastAPI only (dev mode)
> [!WARNING]
> Set environment variable `GRADIO_APP_DISABLED` to `true` to disable Gradio app.
```shell
uvicorn mergeui.main:app --reload --port 8000 --log-level debug
```
#### Gradio app only (dev mode)
```shell
gradio mergeui/web/gradio_app/main.py # with reloading
python mergeui/web/gradio_app/main.py # without reloading
```
#### Bokeh server (dev mode)
```shell
poe bokeh_dev
```
## Testing
This project use pytest for testing, you can run the tests using the following command:
```shell
poe test
```
## Contributing
MergeUI is a new project, and any contribution would make a difference! Whether you find a bug, have valuable feedback
or suggestions, or simply want to get involved, we would love to hear from you!