🚧👷♂️🛠️[WORK IN PROGRESS] SAI2
The opensource AI search platform
## Features
- Integrated support for LLMs, such as OpenAI, Groq, and Claude.
- Native search engine integration, including Google, Bing, DuckDuckGo and SearXNG Search.
- Customizable, visually appealing user interface.
- Shareable and cached search results for enhanced efficiency.
## Setup Search Engine API
Choose your search service
### Search1API
Search1API is a versatile search aggregation service that enables you to perform searches across Google, Bing, and DuckDuckGo, and also retrieve clear content from URLs. [search1api website]( https://go.search1api.com/homepage)
### Bing Search
To use the Bing Web Search API, please visit [this link](https://www.microsoft.com/en-us/bing/apis/bing-web-search-api) to obtain your Bing subscription key.
### Google Search
You have three options for Google Search: you can use the [SearchApi Google Search API](https://www.searchapi.io/) from SearchApi, [Serper Google Search API](https://www.serper.dev) from Serper, or opt for the [Programmable Search Engine](https://developers.google.com/custom-search) provided by Google.
### SearXNG Search
you can host your personal [SearXNG server](https://github.com/searxng/searxng), then you do not need pay for the search api. You just need provide the server address in `SEARXNG_BASE_URL`, plz be sure you enable the json format for the SearXNG server.
## Deployment
### Zeabur
Just click on it
### Docker
Change the environment variables and run the docker
```
docker run -d --name search4all -e OPENAI_API_KEY=sk-XXX -e OPENAI_BASE_URL=https://api.openai.com/v1 -e LLM_MODEL=gpt-3.5-turbo-0125 -e RELATED_QUESTIONS=1 -e SEARCH1API_KEY=XXX -e BACKEND=SEARCH1API -p 8800:8800 docker.io/fatwang2/search4all
```
### Docker-Compose
1. Download the docker-compose file on your mechine
```
wget https://raw.githubusercontent.com/fatwang2/search4all/main/docker-compose.yml
```
2. Change the environment variables in the file
3. Run the docker
```
docker compose up -d
```
### Manual
1. install the requirements.txt
```shell
pip3 install -r requirements.txt
```
2. Set you LLM
```shell
export OPENAI_API_KEY=sk-XXX
OPENAI_BASE_URL=https://api.openai.com/v1
LLM_MODEL=gpt-3.5-turbo-0125
RELATED_QUESTIONS=1
NODE_ENV=production
```
3. Set your key of search
```shell
export SEARCH1API_KEY=YOUR_SEARCH1API_KEY
```
4. Build web
```shell
cd web && npm install && npm run build
```
5. Run server
```shell
BACKEND=SEARCH1API python3 search4all.py
```
## Environment Variable
This project provides some additional configuration items set with environment variables:
| Environment Variable | Required | Description | Example |
| -------------------- | -------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------- |
| `OPENAI_API_KEY` | Yes | This is the API key you apply on the OpenAI account page | `sk-xxxxxx...xxxxxx` |
| `OPENAI_BASE_URL` | No | If you manually configure the OpenAI interface proxy, you can use this configuration item to override the default OpenAI API request base URL | OpenAI: `https://api.openai.com/v1`