General Introduction
Exa & Deepseek Chat App is an open-source smart chat application whose main features include real-time web searching using Exa's API and utilizing the Deepseek R1 Language modeling for intelligent reasoning. Developed by Exa Labs, the application aims to provide users with an efficient and accurate platform for information acquisition and question answering. By combining advanced web search technology and powerful language modeling, users can get the latest web information and accurate reasoning results during the chatting process.
-
Demo address: https://demo.exa.ai/deepseekchat
Function List
- Real-time Web Search: Access to the most up-to-date information on the web via the Exa API ensures that users receive timely and relevant search results.
- intelligent inference: Reasoning about complex problems using the Deepseek R1 language model to provide accurate answers and recommendations.
- open source: Users are free to clone and modify the code to suit different development needs.
- front-end framework: Built on Next.js and TailwindCSS to provide a modern user interface and experience.
- Hosting services: Hosting through Vercel ensures high availability and fast response times for your applications.
- Environment variable configuration: Supports Exa and Fireworks API key configurations for easy customization.
Using Help
Installation process
- clone warehouse::
git clone https://github.com/exa-labs/exa-deepseek-chat.git
cd exa-deepseek-chat
- Installation of dependencies::
npm install
- Setting environment variables::
- Create a
.env
file and add the following:plaintext
EXA_API_KEY=your_exa_api_key
FIREWORKS_API_KEY=your_fireworks_api_key
- Get Exa API key: access Exa dashboard
- Get Fireworks API Key: Access Fireworks Dashboard
- Create a
- Running the development server::
npm run dev
Open your browser and visit http://localhost:3000
Functional operation flow
- Real-time Web Search::
- Enter the query in the chat window and the app will fetch the latest network information via the Exa API and return the results.
- intelligent inference::
- The application uses the Deepseek R1 language model to reason about complex problems and provide accurate answers.
- Custom Development::
- Users can modify the front-end code (Next.js and TailwindCSS) as well as the back-end logic to add new functional modules as needed.
- Hosting and deployment::
- Hosting through Vercel ensures high availability and fast response times for your applications.