Open-Source Skeet Enables Rapid Development of AI Chatbots, Releases Demo with Google's LLM PaLM2 and Vertex AI
Open-Source Skeet Enables Rapid Development of AI Chatbots, Releases Demo with Google's LLM PaLM2 and Vertex AI
ELSOUL LABO B.V. (Headquarters: Amsterdam, Netherlands) has announced that their
development team for the open-source Skeet framework has released a demo
featuring Google's Large Language Model (LLM) PaLM2 and Vertex AI.
Skeet is an open-source, serverless framework built with TypeScript, designed
for full-stack application development on Google Cloud Platform (GCP) and
Firebase.
With Skeet, you can rapidly develop everything from API servers to Web, iOS, and
Android apps using TypeScript. The framework actively incorporates technologies
with a good reputation for developer experience, such as GraphQL and Firestore.
It also facilitates the development of modern applications using AI technologies
like ChatGPT and Vertex AI, as well as blockchain technologies like Solana for
Web3 dApp development.
This release allows users to try out an AI chatbot powered by Google's LLM PaLM2
and Vertex AI via the Skeet demo linked below. You can also test ChatGPT's
GPT-3.5 and GPT-4 from the same demo, enabling you to compare and evaluate which
AI is suitable for different use-cases.
Skeet Demo: https://skeeter.dev/en/
Skeet CLI (GitHub): https://github.com/elsoul/skeet-cli
Google's LLM PaLM2 and Vertex AI
PaLM2 is a Large Language Model (LLM) developed by Google and is now available
on Google Cloud's Vertex AI. It is also the model that powers Bard, another AI
chat service provided by Google.
PaLM2 is trained efficiently using a diverse, multilingual pre-training dataset
and performs well even with fewer tokens or smaller models. The model can be
fine-tuned by providing example prompts, making it easy to create specialized AI
chatbots for various use-cases.
You can try the demo from the link below. Both PaLM2 and Vertex AI, as well as
OpenAI's ChatGPT (GPT-3.5, GPT-4), can be tested simultaneously, allowing you to
compare their respective features.
Skeet Demo: https://skeeter.dev/en/
The demo can be run on your own PC or cloud environment in just 5 minutes using
Skeet CLI.
Skeet CLI (GitHub): https://github.com/elsoul/skeet-cli
Skeet is developed as an open-source project aimed at reducing development and
maintenance costs and improving the developer experience worldwide.
For more details, please refer to the official documentation.
Skeet Documentation: https://skeet.dev/en/
