Post

How to get URL link on X (Twitter) App

On the Twitter thread, click on or icon on the bottom
Click again on or Share Via icon
Click on Copy Link to Tweet
Paste it above and click "Unroll Thread"!
More info at Twitter Help

Shubham Saboo

Mar 20 • 7 tweets • 4 min read • Read on X

Scrolly

5 AI agent frameworks to build multi-agent applications.

100% opensource.

They aren't LangChain, Crew AI, or OpenAI Agents SDK.

1. Motia is an AI agent framework is built for Software Engineers. Build agents in Python, TypeScript, JavaScript or Ruby.

github.com/MotiaDev/motia

2. Agno is a lightweight Python library for building Multimodal Agents with memory, knowledge, and tools.

• ~10,000x faster than LangGraph
• Build teams of specialized agents
• Make Agents respond with structured data

github.com/agno-agi/agno

3. AWS Multi-Agent Orchestrator is a flexible framework to build multi-agents AI applications.

4. • Routes queries to the most suitable agent
• Fully implemented in both Python and TypeScript
• Variety of ready-to-use agents

github.com/awslabs/multi-…

4. PydanticAI is a Python based agent framework to build production grade applications.

• Model-agnostic
• Python-centric Design
• Structured outputs

github.com/pydantic/pydan…

5. AutoAgent creates and deploys fully-automated AI agents with no-code. Build multi-agent teams and agentic RAG apps from simple prompts.

It is equipped with a native self-managing vector database, outperforms industry-leading solutions like LangChain.

github.com/HKUDS/AutoAgent

I have created 50+ AI Agents and RAG tutorials, 100% free and opensource.

Two simple steps to get started:
1. Subscribe to Unwind AI (for free): theunwindai.com
2. Star the repo: github.com/Shubhamsaboo/a…

New AI Agents and RAG tutorials added every week.

https://x.com/Saboo_Shubham_/status/1902547400893591584

If you find this useful, RT to share it with your friends.

Don't forget to follow me @Saboo_Shubham_ for more such LLM tips and AI Agent, RAG tutorials.

https://x.com/Saboo_Shubham_/status/1902547400893591584

• • •

Missing some Tweet in this thread? You can try to force a refresh

This Thread may be Removed Anytime!

Twitter may remove this content at anytime! Save it as PDF for later use!

More from @Saboo_Shubham_

Shubham Saboo

@Saboo_Shubham_

Mar 20

OpenAI just released Voice AI Agents with GPT-4o Audio model.

You can now build amazing voice AI Agents using OpenAI Agents SDK in a few lines of Python code.

Quickstart code to build Voice AI Agents with function calling in Python.

You can play with the new voice AI model at for free. openAI.fm

Read 5 tweets

Shubham Saboo

@Saboo_Shubham_

Mar 11

OpenAI just released Agents SDK for building multi-agent apps in Python.

It comes with 3 key features:
1. Handoffs: Allows an agent to delegate tasks to another agent
2. Guardrails: Safety checks for input & output validation
3. Tracing: Built-in visual tracking of agent runs.

Handoffs allow an agent to delegate tasks to another agent.

It's particularly useful in scenarios where different agents specialize in distinct areas.

Guardrails validate user input and agent output, preventing misuse and saving costs. A fast, cheap model can block unwanted requests before reaching an expensive one.

Types of guardrails:
• Input guardrails check user input.
• Output guardrails validate the agent’s response.

Read 7 tweets

Shubham Saboo

@Saboo_Shubham_

Mar 11

Opensource alternative to Manus AI Agent is blowing up on GitHub.

OWL is an Autonomous AI Agent framework that can research, browse and code with multi-agent collaboration.

Works with Claude Sonnet 3.7, DeepSeek, GPT-4o and even local LLMs with Ollama.

100% Opensource.

OWL ranks 1 on the GAIA Benchmark among open-source projects with a 58.18 avg score.

Try it out yourself: github.com/camel-ai/owl

Read 5 tweets

Shubham Saboo

@Saboo_Shubham_

Feb 22

Train your own DeepSeek R1 like reasoning model with just 5GB of RAM.

It enables training custom reasoning models with 10x longer context while using 90% less VRAM.

100% Opensource.

Find all the awesome LLM Apps with AI Agents and RAG in the following Github Repo.

P.S: Don't forget to star the repo to show your support 🌟

github.com/Shubhamsaboo/a…

Read 6 tweets

Shubham Saboo

@Saboo_Shubham_

Feb 8

AI Agents can now do Object Detection

AI agents can now detect objects through pure reasoning - just provide text prompt & image.

Zero training data needed. No labelling required.

Find all the awesome LLM Apps with AI Agents and RAG in the following Github Repo.

P.S: Don't forget to star the repo to show your support 🌟

github.com/Shubhamsaboo/a…

50+ Step-by-step tutorials of LLM apps with AI Agents and RAG.

P.S: Don't forget to subscribe for FREE to access future tutorials.

theunwindai.com

Read 4 tweets

Shubham Saboo

@Saboo_Shubham_

Feb 5

4 alternatives to OpenAI’s $200/month Deep Research AI Agent.

100% Opensource and $0 subscription fees.

1. DeepResearch by Jina AI

It uses Gemini Flash, Brave/DuckDuckGo, and Jina Reader, to mimic OpenAI's agentic search, read, and reasoning process.

100% Opensource.

github.com/jina-ai/node-D…

2. Open Deep Research by Firecrawl

It uses Firecrawl search + extract along with the AI SDK for reasoning with any LLM, instead of using a fine-tuned version of o3.

github.com/nickscamara/op…