2 weeks to go until the fal x @sequoia 72 Hour Video Hackathon.
Build the future of video with the best AI models in the world.
July 17-19.
Apply now: https://t.co/JFeWqdPL7R
New free model in Cline! Laguna M.1 by Poolside.
Speedy 225B total parameter model with 256k context, built for agentic coding and long-horizon work.
Use the model id: poolside/laguna-m.1:free
Interesting new optional skill for Hermes Agent called unbroker. I made a quick video showing how to install and set it up, and what results you might see when you run it.
So what is unbroker? Simply, it finds where data brokers have your personal info exposed online and files the removal requests for you.
We all know our data gets stored and sold. A lot of brokers are legally required to delete it if you ask, but doing that across dozens of sites by hand is miserable. Hermes Agent has it as a built-in security skill, so I just let my agent run the whole thing.
How it went:
- Set up browser automation (used Browserbase, just an API key + project ID in hermes tools).
- Pointed Hermes at the GitHub, said "install this skill." Done.
- Gave explicit consent, which it requires before doing anything, plus a quick intake: legal name, past names, cities, emails.
- "Use the unbroker skill to remove my data."
It spun up sub-agent swarms and scanned 51 broker sites. Real results are in the video.
The best part is it's built to loop. It drafts the opt-out emails, or if you connect your email it sends them hands-off, then schedules rechecks and logs everything. Set it once and it keeps your data clean over time.
Kind of wild that one skill and a couple prompts can check 50+ broker sites for you. Underrated use of agents. Let me know your thoughts!
WaveLander: A Generalizable Hierarchical Control Framework for UAV Landing on Wave-Disturbed Platforms via Reinforcement Learning
Chun-Kit Li, Iok Long Sit, Ming Fung Siu, Ka Yu Kui, Hin Wang Lin, Pengyu Wang, Ling Shi
https://t.co/XS1F7mqZkF [𝚌𝚜.𝚁𝙾]
Open model usage has gone from 10% of AI tokens to 30% in a year.
The shift to open, modular AI is here to stay. Our founders on what's driving it, and where it goes from here.
Enterprise MCP just got a major upgrade.
MCP’s Enterprise-Managed Authorization extension is now stable, enabling organizations to centrally manage MCP server access through identity providers instead of handling individual OAuth approvals for every connection.
A key step toward bringing agentic workflows into enterprise environments.
Read more from @TheNewStack: https://t.co/8x7SIRzB1I
Great AI videos don't start with rendering.
They start with direction.
The real creative skill isn't just generating videos anymore—it's directing them through words.
@VettedPodcast@GoodTroubleShow Think it's a great idea personally, they do it in politics and other areas, perhaps make a monetary wager between the two the loser donates to a nonprofit of the winners choosing, let the audience decide? Just a thought 🤷♂️
I challenge @GoodTroubleShow to a friendly debate about the “UFO Legacy Program”.
What do you say Matt?
You called me out. Now show up.
LIVESTREAMED, no edits.
Mediator of your choosing.
Day and time of your choosing.
Everything in your favor.
I'm interested in how you're all running Hermes day to day. drop your setup below, I'm mapping what the community reaches for.
I'm mostly curious about:
- model: your daily driver, plus MoA or a local model if you run one
- memory: built-in, an Obsidian vault, or another layer
- interface: TUI, Desktop, or a Messaging gateway
- orchestration: kanban, delegate_task, subagents in tmux, /goal
- the skills or MCP servers you'd miss if they were gone
no setup is too small. I'll gather the common patterns and share them back.
AgenticDataBench
344 realistic tasks across 15 domains.
97 datasets and 433 data science skills.
Fine-grained labels to rigorously evaluate LLM data agents.
Sensorless Four-Channel Control Architecture Using Inverse Dynamics Modeling for Human-Scale Bilateral Teleoperation
Amir Noohian, Dylan Miller, Justin Valentine, Alan Lynch, Martin Jagersand
https://t.co/eDKLun3Dlc [𝚌𝚜.𝚁𝙾]
Jeff Bezos predicts AI will dramatically accelerate invention and make engineers superhumanly productive.
- His AI startup, Prometheus, is valued at $41 billion and is developing an "Artificial General Engineer."
- Today's "dream-to-build" cycle can take 10 years. Bezos believes AI could shrink that to 5 years, then 3 and eventually just 1 year.
- Instead of training only on internet text, next generation engineering AI models will learn from physics, simulations, manufacturing processes and real-world engineering data.
This could dramatically accelerate the development of everything from jet engines to advanced robots, enabling engineers to build breakthrough technologies faster than ever before.
Big things come in small packages – and MiniCPM5‑1B proves it once again. 🥳
A developer in our community @OrlandoLiuqn built EdgeHome Harness – a full‑stack Rust framework that turns a lightweight 1B model into a reliable, deterministic controller for smart home and vertical scenarios:
⚡ Runs on just 2GB RAM
MiniCPM5‑1B (688MB, ~1GB runtime) + Rust Agent Harness (25MB‑64MB) – the entire stack fits comfortably in ultra‑constrained edge devices.
🧠 Model proposes, Harness disposes
User command → MiniCPM5 generates candidate JSON → Rust validates, plans, executes, and traces.
The model never touches devices directly – every candidate must pass room/device existence, capability checks, and policy gates (auto‑execute, confirm, or fallback).
🛡️ Handles common 1B‑model issues
Output Governor handles repetition, rambling, and malformed JSON. Rust manages short‑term memory (last_target, aliases) to avoid context bloat. Device Registry + Policy Gate enforce deterministic security boundaries.
📊 Fully observable & replayable
Trace / Replay / Eval make every failure reproducible and measurable – no more tuning by instinct.
This isn’t a commercial gateway – it’s an engineering prototype that proves: small models don’t need to be omnipotent. With the right harness, MiniCPM5‑1B works reliably at the edge. It’s about redefining the model‑system boundary, not cramming cloud AI into tiny boxes.
🔗 Original post: https://t.co/ZgGymo4Fvw
GitHub: https://t.co/0qtAnUGEhL
Meet WebBrain: An Open-Source, Local-First AI Browser Agent That Reads Pages and Automates Tasks in Chrome and Firefox
WebBrain lives inside your browser and can run entirely on your own local model — no cloud, no account, no data leaving your machine.
Most "AI browser agents" are a chat box that pastes your page into someone else's server. That's not an agent that lives where you browse — and WebBrain draws a very clear line between the two.
It's an open-source (MIT), local-first browser agent for Chrome and Firefox. It runs inside your existing authenticated session, on a model you pick — so with llama.cpp or Ollama, nothing leaves your machine.
Here's what's actually interesting:
→ Two modes, cleanly separated. Ask reads the page (read-only, content scripts). Act clicks and types through the Chrome DevTools Protocol (chrome.debugger) — trusted input events that modern sites honor, reaching cross-origin iframes and shadow DOM.
→ UI-first by design. For anything that submits, sends, or buys, it drives the visible UI and refuses to hit REST/GraphQL endpoints directly. It starts read-only and asks before consequential actions.
→ Bring any model. llama.cpp, Ollama, LM Studio, vLLM — or OpenAI, Claude, Gemini, DeepSeek, Groq, OpenRouter. Recommended local: Qwen 3.6 35B (Qwen3.6-35B-A3B), which beat Gemma 4 on the project's screenshot benchmark.
→ Tuned for cost and privacy. Token-conscious screenshots, oldest-first context trimming, a dedicated vision model, 40+ tools (~20 in Compact mode). No telemetry. No accounts.
Full analysis: https://t.co/ITKYPYx97u
GitHub Repo: https://t.co/3xHZ9nhmJl
Chrome Extension: https://t.co/BhbllJCwqF
Firefox Add-on: https://t.co/uX4pkm9NXv
Portal: https://t.co/DH0J8kOPxY
Here is something completely new you can do in one shot with Runway.
This entire video was generated from a single long audio file.
Runway analyzed the file, read the audio transcription, and generated one coherent video to bring it all together visually.