Skip to content

Latest commit

 

History

History
136 lines (94 loc) · 11.8 KB

README-EN.md

File metadata and controls

136 lines (94 loc) · 11.8 KB
Raw

简体中文 · English

User Communication

Telegram Channel

Buy me a coffee

Buy Me A Coffee

Version Updates

  • V0.2.6, 20240425, support the searxng search service, support the moonshot API in stream mode
  • V0.2.5, 20240425, open source the code for the search api
  • V0.2.4, 20240424, support for Groq in Cloudflare Worker
  • V0.2.3, 20240423, support for Azure OpenAI in Cloudflare Worker. It also introduces the ability to use an authorization code and customize the user's request key.
  • V0.2.2, 20240420, support Moonshot API on unstream mode
  • V0.2.1, 20240310, supports Google, Bing, Duckduckgo, Search1API for news-type searches; supports adjusting the number of search results via the MAX_RESULTS environment variable; supports adjusting the number of in-depth searches desired via the CRAWL_RESULTS environment variable.
  • V0.2.0，20240310，Optimized openai.js, cloudflare worker version, really faster this time!

For more historical updates, please see Version History

S2A

Help your LLM API support networking, search, news, web page summarization, has supported OpenAI, Gemini, Moonshot, the big model will be based on your input to determine whether the network, not every time the network search, do not need to install any plug-ins, do not need to replace the key, directly in your commonly used OpenAI/Gemini three-way client replacement of custom You can directly replace the customized address in your usual OpenAI/Gemini three-way client, and also support self-deployment, which will not affect the use of other functions, such as drawing, voice, etc.

效果示例 效果示例
效果示例 效果示例

Features

Model Features Stream Deployments
OpenAI search, news, crawler stream, unstream Zeabur, Local deployment, Cloudflare Worker, Vercel
Azure OpenAI search, news, crawler stream, unstream Cloudflare Worker
Groq search, news, crawler stream, unstream Cloudflare Worker
Gemini search stream, unstream Cloudflare Worker
Moonshot search, news, crawler stream(only on cf), unstream Zeabur, Local deployment, Cloudflare Worker(stream), Vercel

Usage

Replace the custom domain in any client with the following address

image

Deployment

Zeabur

Click the button for one-click deployment, switched on your own environment variables

Deploy on Zeabur

To keep the project updated, it is recommended to fork this repository first, then deploy your branch through Zeabur

Deployed on Zeabur

Local Deployment

  1. Clone the repository locally
git clone https://github.com/fatwang2/search2ai
  1. Copy .env.template as .env, configure environment variables
  2. Enter the api directory, run the program, and display the log in real-time
cd api && nohup node index.js > output.log 2>&1 & tail -f output.log
  1. Port 3014, the complete address after concatenation is as follows, can be configured according to the client's requirements for the apibase address (if https is required, need to use nginx for reverse proxy, many tutorials online)
http://localhost:3014/v1/chat/completions

Cloudflare worker

  1. Copy the code of search2openai.js, or search2gemini.js, or search2groq.js, no modifications needed! Deploy in cloudflare's worker, after going online, the worker's address can be used as your interface call's custom domain address, note the concatenation, worker address only represents the part before v1
  2. Configure variables in the worker（only openai） Effect Example
  3. Configure triggers - custom domain in the worker, direct access to the worker's address in China might have issues, need to replace with custom domain Alt text

Vercel

Special note: Vercel project does not support streaming output and has a 10s response limit, actual user experience is poor, released mainly for experts to pull request

One-click deployment

Deploy with Vercel

To ensure updates, you can also first fork this project and then deploy it on Vercel yourself

Environment Variables

This project provides some additional configuration options, which can be set through environment variables:

Environment Variable Required Description Example
SEARCH_SERVICE Yes Your search service. The key of the service you choose needs to be configured. search1api, google, bing, serpapi, serper, duckduckgo, searxng
APIBASE No Third-party proxy address. https://api.openai.com, https://api.moonshot.cn, https://api.groq.com/openai
MAX_RESULTS Yes Number of search results. 10
CRAWL_RESULTS No The number of deep searches (retrieve the main text of the webpage after searching). Currently only supports search1api, deep search will be slow. 1
SEARCH1API_KEY Conditional Required if search1api is selected. My own fast and cheap search service. Apply at https://search1api.com. xxx
BING_KEY Conditional Required if Bing search is selected. Please search for the tutorial yourself. Apply at https://search2ai.online/bing. xxx
GOOGLE_CX Conditional Required if Google search is selected. Search engine ID. Please search for the tutorial yourself. Apply at https://search2ai.online/googlecx. xxx
GOOGLE_KEY Conditional Required if Google search is selected. API key. Apply at https://search2ai.online/googlekey. xxx
SERPAPI_KEY Conditional Required if serpapi is selected. Free 100 times/month. Register at https://search2ai.online/serpapi. xxx
SERPER_KEY Conditional Required if serper is selected. Free 2500 times for 6 months. Register at https://search2ai.online/serper. xxx
SEARXNG_BASE_URL Conditional Required if searxng is selected. Fill in the domain name of the self-built searXNG service, refer to this repo https://github.com/searxng/searxng, plz open the json mode https://search.xxx.xxx
OPENAI_TYPE No OpenAI provider source, default is openai openai, azure
RESOURCE_NAME Conditional Required if azure is selected xxxx
DEPLOY_NAME Conditional Required if azure is selected gpt-35-turbo
API_VERSION Conditional Required if azure is selected 2024-02-15-preview
AZURE_API_KEY Conditional Required if azure is selected xxxx
AUTH_KEYS No If you want users to define a separate authorization code as a key when making requests, you need to fill this in. Required if azure is selected 000,1111,2222
OPENAI_API_KEY No If you want users to define a separate authorization code as a key when requesting openai, you need to fill this in sk-xxx

Future Iterations

  • Fix streaming output issues in Vercel project
  • Improve the speed of streaming output
  • Support more vertical searches