- unwind ai
- Posts
- Drag & Drop Canvas for AI Agents
Drag & Drop Canvas for AI Agents
PLUS: New Gemini 1.5 Pro & Flash, OpenAI Advanced Voice mode rolled out
Shubham Saboo & Gargi Gupta
September 25, 2024
In partnership with
FREE AI & ChatGPT Masterclass to automate 50% of your workflow
More than 300 Million people use AI across the globe, but just the top 1% know the right ones for the right use-cases.
Join this free masterclass on AI tools that will teach you the 25 most useful AI tools on the internet – that too for $0 (they have 100 free seats only!)
This masterclass will teach you how to:
Build business strategies & solve problems like a pro
Write content for emails, socials & more in minutes
Build AI assistants & custom bots in minutes
Research 10x faster, do more in less time & make your life easier
You’ll wish you knew about this FREE AI masterclass sooner 😉
Today’s top AI Highlights:
Build AI Agent workflows with drag-and-drop low-code canvas hosted locally
Google releases new versions of Gemini 1.5 Pro and Flash models
OpenAI starts rolling out Advanced Voice Mode to all Plus and Team users
Anthropic might get valued at $40 billion
MacOS native app to chat with opensource models right on your desktop
& so much more!
Read time: 3 mins
AI Tutorials
The tech world is evolving so fast that staying up-to-speed is overwhelming. How about multiple AI agents doing that research for you from the most dynamic yet cluttered source of top tech stories, Hacker News?
In this tutorial, we’ll show you how to build an AI-powered multi-agent researcher to research top stories on HackerNews, generating blog posts, reports, and social media content, all autonomously and in just 15 lines of Python code.
We share hands-on tutorials like this 2-3 times a week, designed to help you stay ahead in the world of AI. If you're serious about levelling up your AI skills and staying ahead of the curve, subscribe now and be the first to access our latest tutorials.
🎁 Bonus worth $50 💵
Latest Developments
Google has released two production-ready models, Gemini-1.5-Pro-002 and Gemini-1.5-Flash-002, along with reduced costs, increased rate limits, and faster output speeds. These updates are available via the Google AI Studio and the Gemini API. The models also feature significant improvements in math, long-context understanding, and vision-related tasks, alongside refined filter settings and concise responses based on developer feedback.
Key Highlights:
Better and faster - Both Gemini 1.5 Pro and Flash performed better on MMLU-Pro and benchmarks for math, Python code generation, and vision-related tasks. Further, Gemini 1.5 Flash’s output tokens per second are 2x more than before, making it optimal for apps requiring real-time responses.
More concise outputs - Both models now have a more concise style in response to developer feedback to make these models easier to use and reduce costs. The output length of the new models is ~5-20% shorter than previous models.
API prices further reduced - Google is reducing the input tokens price by 64%, output tokens price by 52%, and incremental cached tokens by 64% for Gemini 1.5 Pro, effective October 1st, 2024, on prompts <128K tokens.
Increased rate limits - Rate limits have been doubled for Gemini 1.5 Flash (2,000 RPM) and tripled for Gemini 1.5 Pro (1,000 RPM), so you can handle more requests.
Try today - You can try all the new models in Google AI Studio and Vertex AI today. Trust us, you’ll be blown by the speed of Gemini 1.5 Flash's responses to multimodal inputs!
A low-code drag-and-drop interface to build complex AI agent workflows, hosted locally - isn’t it like every exciting thing put together? AutoGPT has just released its platform that lets you create, deploy, and manage continuous AI agents in a single intuitive canvas. These AI agents run continuously on your behalf to automate complex tasks. AutoGPT offers a user-friendly, low-code interface where workflows are constructed by simply dragging, dropping, and connecting blocks. It allows for both local and cloud-based deployment.
Key Highlights:
Intuitive Build Interface - Build, modify, and optimize your automation workflows with ease. You build your agent by connecting blocks, where each block performs a single action.
Ready-to-Use Agents - Don't want to build? Simply select from AutoGPT’s library of pre-configured agents and put them to work immediately. If you want to create custom blocks, read this guide.
Flexible Deployment Options - Deploy your AI agents locally for full control or utilize cloud hosting to ensure they operate continuously and reliably.
24/7 Autonomous Agents - Set up your agents to automatically execute tasks round-the-clock, whether for data analysis, content generation, or customer outreach.
Deployment Control and Monitoring - Manage and monitor your agents efficiently with comprehensive tools that track performance and facilitate real-time adjustments.
Get started now - To self-host, you can download from this GitHub repo and follow the instructions to set it up. Here’s a quick demo of what the UI looks like and how you can start building workflows on it. For cloud-hosted beta, join the waitlist here.
Quick Bites
OpenAI has (finally!) started rolling out the Advanced Voice Mode to all Plus and Team users in the ChatGPT app starting today and the rollout will be completed in a week. It seems they tried to compensate for the wait by adding Custom Instructions, Memory, five new voices, and improved accents in the Voice Mode. It can also say “Sorry I’m late” in over 50 languages (was it necessary though?!). You will see a notification in the app when you have access to it.
YouTube is expanding access to its conversational AI tool in the app for all U.S. YouTube Premium members with Android devices. While you’re watching a video on YouTube, the conversational AI tool lets you interact with AI to learn more about the content.
Google is rolling out a super cool AI-powered video presets feature in the Google Photos app. These presets can automatically trim the video, adjust lighting, control speed, or apply effects like dynamic motion tracking of the main subject, zooming in the main action, or applying slow-mo with just a few taps.
Anthropic is reportedly in early talks to raise new funding at a valuation of up to $40 billion. The AI startup has previously raised $7.6 billion, with major investors including Amazon, which could participate in the new round.
Tools of the Trade
HuggingChat: A native macOS app that brings powerful open-source language models like Qwen and Llama straight to your desktop - with markdown support, web browsing, code syntax highlighting, and much more!
Embedded Recording SDK by Velt: Quickly add AI-powered voice, video, and screen recording, like Loom, into your product. It also generates automatic summaries and transcriptions, making it easy to keep users engaged within your platform.
JSON For You: AI tool for visualizing, formatting, comparing, and processing JSON data efficiently using both graph and table views. It supports local data processing, JSON validation, CSV conversion, and specialized comparisons.
Awesome LLM Apps: Build awesome LLM apps using RAG to interact with data sources like GitHub, Gmail, PDFs, and YouTube videos through simple text. These apps will let you retrieve information, engage in chat, and extract insights directly from content on these platforms.
Hot Takes
The criticisms of why LLM’s can’t reason are disturbingly relevant to people as well. ~
Marc AndreessenIf AGI is achievable, the very first people who will know it will be traders who will suddenly find all of their good strategies & trades don’t work anymore, and that an unknown firm is on the winning side of all of them. ~
Ethan Mollick
Meme of the Day
That’s all for today! See you tomorrow with more such AI-filled content.
🎁 Bonus worth $50 💵
Share this newsletter on your social channels and tag Unwind AI (X, LinkedIn, Threads, Facebook) to get AI resource pack worth $50 for FREE. Valid for limited time only!
PS: We curate this AI newsletter every day for FREE, your support is what keeps us going. If you find value in what you read, share it with at least one, two (or 20) of your friends 😉
Reply