.jpg)
Eden AI is excited to announce a raft of new features for its custom chatbot builder, empowering developers to create even more sophisticated and engaging chat experiences.
Eden AI's Chatbot solution using RAG is a versatile workflow developed by Eden AI that empowers users to create custom chatbots on their own data or business-specific information with any AI model from a wide range of LLMs available on the market: OpenAI GPT 4, Cohere Command, Google Cloud PaLM2, Meta Llama2, and more.
Your chatbot can be integrated into a website or in Discord to allow users to ask questions and receive responses based on the data the chatbot has been trained on. The repository on GitHub contains the source code for using and displaying your Chatbot in a website, with branches for the unframed source code and the embed code.
Eden AI’s Custom Chatbot addresses limitations by facilitating data integration and training in multiple programming languages. It has broad applications across industries, making it a versatile tool for businesses, students, content creators, and researchers to train chatbots with their own data.
At the core of the update lies a shift in the underlying LLMs used for chatbot interactions. The ask_llm endpoint now uses chat-specialized models like GPT-4, Claude 3, and Cohere R, these models are specifically designed for conversational scenarios ensuring your chatbot delivers more natural and relevant responses, and for RAG.
Additionally, you can optimize the conversation by customizing settings like maximum token limit and temperature.
Building rich, multi-turn conversations just got easier. Eden AI now allows you to save multiple conversations per project. You can choose to import existing conversation history or create new ones and assign them unique IDs for your chatbot to reference. This enables the chatbot to maintain context and build upon previous interactions, leading to a more personalized and engaging experience for users.
For developers seeking greater control over their data, Eden AI introduces custom database integration. You can now specify your own database provider resources (key and account details) to store chatbot data within your preferred platforms like Qdrant or Supabase. This ensures your data remains securely housed in your chosen environment.
The application interface for managing conversation chunks has been significantly enhanced. You can now easily view the full content of each chunk, providing a clearer understanding of your chatbot's conversation history.
Uploading various file formats is now a breeze with the new unified upload endpoint. The system supports audio, PDF, XML, and CSV files. Additionally, you can track the upload progress, receiving clear indications of success or failure. For added convenience, the option to delete all uploaded content associated with a specific file is also available.
Project creation has become more customizable with the ability to select preferred providers for PDF parsing (OCR) and audio transcription (speech-to-text). This allows you to tailor the system to your specific needs and data sources. Furthermore, you have finer control over data management by defining chunk sizes and separators during project creation.
Eden AI empowers you to personalize your chatbot's behavior. You can define custom prompts for the bot or leverage the new chatbot_global_action message system. This enables the chatbot to perform specific actions based on pre-defined prompts, offering a more interactive and dynamic user experience.
To start using LLMs for your Custom Chatbot on Eden AI, you'll need to create an account for free. Then, you'll be able to get your API key directly from the homepage with free credits offered by Eden AI.
With these powerful new features, Eden AI's custom chatbot builder empowers developers to create next-generation conversational experiences that are both intelligent and engaging.
You can directly start building now. If you have any questions, feel free to schedule a call with us!
Get startedContact salesDrop your email and we'll get back to you ASAP to answer any questions you have or just to say hi —we promise not to spam you!
-p-500.webp){kind=small}
-p-500.webp){kind=small}
