Letโs build a Local AI Agent with Knowledge and Storage Using Agno ๐
- 100% Local and Open Source
- @AgnoAgi to implement the AI agent
- Agno Agent UI as an intuitive Agent interface
Link to our FREE guide in the comments.
Turn any GitHub repo into a diagram! ๐คฏ
This helps to understand complex projects.
Just replace 'hub' with 'diagram'.
You can also download the diagram as a PNG or Mermaid file.
100% Open Source
Link to the GitHub Repo in the comments.
Perplexity has launched its first AI web browser, called Comet. ๐ง
This move puts the AI startup in direct competition with Google Search. The battle for search dominance has begun.
More below ๐๐ฝ
Elon Muskโs xAI has released the fourth version of Grok. ๐คฏ
According to xAI, itโs โthe most intelligent model in the world.โ Elon Musk also said that Grok 4 is better than any PhD student in any field.
The details ๐๐ฝ
The Chinese AI startup Moonshot AI has released its powerful Large Language Model (LLM), Kimi K2, as open-source on GitHub. ๐ฅ
It performs similarly to leading closed-source models on some benchmarks.
More below ๐๐ฝ
What is Tool Calling? ๐ง
Tool calling refers to the ability of LLMs to interact with external tools, APIs, or systems to improve their functionality.
Hereโs how it works:
Pytest for LLM Apps ๐ฅ
DeepEval from Confident AI is an LLM evaluation framework designed for testing LLM systems. It works like Pytest but is specifically made for unit testing LLM outputs.
With DeepEval, you can easily find the best models, prompts, and architecture to enhance your RAG pipeline and agentic workflows.
100% Open-Source
Link to the Github Repo in the comments.
Train, fine-tune, and evaluate LLMs on your own machine! ๐ฅ
Transformer Lab makes it possible. The app lets you download LLMs with one click, use different inference engines, and has built-in logging and more.
100% Open-Source
Link to the GitHub Repo in the comments.
Fine-tune LLMs like Llama or Mistral using 80% less memory ๐คฏ
Unsloth makes it possible.
And the BEST part: You can try it out yourself right away.
๐๐ฝ 100% Open Source
Link to the GitHub Repo in the comments.
AG-UI completes the agent protocol stack ๐ฅ
- MCP connects agents to tools.
- A2A enables agents to communicate with other agents.
- AG-UI (@CopilotKit) connects agents to users
Letโs build an AI Agent to Chat with Financial Charts ๐
- 100% Local and Open Source
- Agno to implement the AI agent
- Chainlit as an intuitive UI
Step-by-Step Guide (Python Code on GitHub) ๐๐ฝ
https://t.co/dZkbxaqnbw