AI This Week: Unleashing the Power of AI - From Autonomous Assistants to Multilingual Models

Top Repo: AI Assistants

PhiData is a new framework for building Autonomous Assistants, also known as Agents. These agents are equipped with long-term memory, contextual knowledge, and the ability to take actions using function calling.

The problem PhiData addresses is that Language Learning Models (LLMs) have limited context and cannot take actions. The solution is to add memory, knowledge, and tools to these models.

• Memory: Stores chat history in a database and enables LLMs to have long-term conversations.
• Knowledge: Stores information in a vector database and provides LLMs with business context.
• Tools: Enable LLMs to take actions like pulling data from an API, sending emails, or querying a database.

The process of using PhiData involves three steps:

1. Create an Assistant
2. Add Tools (functions), Knowledge (vectordb), and Storage (database)
3. Serve using Streamlit, FastApi, or Django to build your AI application

To install PhiData, use the command !pip install -U phidata.

Trending Signals

• Open Source: Mistral releases ‘Mistral-Finetune’, the official repo to fine-tune Mistral open-source models using Lora.
• Finance: GPT-4 can help pick stocks, beating humans and other machine learning models trained for finance.
• Multilingual: Cohere releases Aya 23, a powerful open weight multilingual model with 101 languages.
• RAG: LlamaIndex introduces RAGApp, a no-code interface to configure a RAG chatbot, as simple as GPTs.
• Theory: Yann LeCun claims there’s no such thing as AGI: Cats and dogs are more intelligent than current AI systems.

Subscribe to Newsletter : https://lnkd.in/guxfrUSM