Lucida from ByteDance Cool project! It transforms indoor video into editable 3D scenes; reconstructs objects as simulation-ready assets - VLM-based object detection and referring descriptions. - good at pose alignment - high-fidelity real-to-sim digital twins. - multi-view masks, boxes, partial point clouds - Seed3D 2.0 for image-to-3D asset gen https://t.co/uUlcInA2XL

Here’s what launched this week: — Gemini 3.5 Transcribe, our most precise speech-to-text model yet, designed to deliver intelligent transcriptions — Gemini Omni 1.1 Flash, bringing expanded creative capabilities and controls for video generation and editing — @GeminiApp’s Live experience, moving beyond conversation to complex tasks with new features like Daily Brief, Gemini Spark, Personal Intelligence, and @Gmail inbox management — Expert Intelligence, a new cross-@Google initiative that lets you engage with and combine insights from trusted sources, starting with eligible @GooglePlay ebooks in @Gemini_Notebook

there needs to be a communication protocol between agents I just want your agent to talk to my agent. not your agent send me a slack dm that I copy paste to my agent.

My workflows changed when I started treating each Grok Bot like an application Client: chat Server: @bot DB: anything (gh, sheets, SaaS) I've been white-pilled on "agents for everything" - thin client, thick server

There is a popular twitterpinion going around that companies should stop building agents and only expose MCPs/CLIs. I agree the old way of clicking through a UI is going away but this take is short sighted and misses nuance. It is almost impossible to expose the full strength of a product through a flat list of tool calls. The API calls behind a real workflow need careful orchestration. And authz is incredibly hard in a headless interface. Companies that want to take advantage of this shift need all three: 1. An MCP their users can plug into any agent 2. An in-product agent: a carefully curated experience the company controls, owns, and learns from. It becomes a new way of using the product. 3. The product agent exposed through the MCP. Users get the company's agent through their agent of choice. Simple requests like "pull up my subscriptions on stripe" go through the MCP tools. Complex, curated experiences go through the company's agent (accessed via MCP as well). Everything should be reachable through MCP, so your agent can talk to my agent. But expecting a general-purpose agent to orchestrate a complex chain of tool calls in the right order every time, is laughable. The orchestration is the product. The problem is that most companies haven't cracked that yet... so its easy to say: give me all the tools. That's like saying let me use your REST API directly. We just haven't seen the good agentic experiences yet and making this take the standard means we never will.

Generate with any model on Runway, then take it to delivery spec with Ruby. Seedance 2.5, Gen-4.5, MiniMax H3, whatever you use, converts straight to 16-bit EXR or 10 and 12-bit ProRes and HEVC. Try it now at the link below. https://t.co/Qb1wKAO0sP

everyone is looking at grok bot like it is five new employees the more interesting part is that they are not five isolated employees they share one persistent cloud environment: the same browser sessions, terminal, filesystem and authenticated tools - while a chief of staff can route work between them without waiting for you to open the laptop that turns one login into infrastructure a research bot can collect leads, pass them to a writer, hand the result to a designer and let an ops bot update the workflow all inside the same running machine the upside is obvious: less copying, fewer handoffs and work that continues while you sleep the blast radius is just as obvious: - authenticate only the services the whole agent fleet is allowed to reach - give every bot a narrow role description, because that description becomes its place in the org chart - keep irreversible actions behind human approval: sending, publishing, spending and production changes - store skills, configuration and long-term state inside `/workspace`, because the rest of the VM may disappear during an update - treat shared sessions as shared authority, not as separate bot identities the same architecture that makes grok bot useful is what makes it dangerous to configure casually my take: the breakthrough is not that you can hire five AI workers it is that they can operate like one persistent company inside your real tools and your job changes from prompting every task to designing the permissions, boundaries and handoffs the full chief-of-staff setup, session handoff model, approval matrix and `/workspace` survival strategy are mapped out in the article ↓

MuScriptor, our music transcription model, can now make sheet music and tabs! We export PDFs and editable MusicXML. Under the hood, we generate these from MIDI using @musescore + some custom preprocessing. https://t.co/zWHqbR0zmJ

This is my new dev setup btw - whenever claude is cooking I'm gobbling up the infinite slop in VR https://t.co/y2pgQcPoUn

Releasing gpt4-x-vicuna-13b, a vicuna base trained on over 180k censor-free GPT-4 outputs. 5 epochs. Datasets by Nous Research, @teknium, WizardLM, and GPT-4-LLM Next model: cleaning vicuna to fully remove OAI censorship. compute sponsor: Redmond AI https://t.co/tP3pcEzkKr

Today marks 6 months of Hermes Agent 6 months ago, all of our workflows looked very different Today, thousands of us use Hermes to build, automate, research, code, run agents, and manage real workflows every day I think we forget how fast this happened For me, Hermes changed almost everything about how I work So today is a good day to say thank you. Thank you @NousResearch, @Teknium, every contributor, and this entire community for building something that became much more than just another AI tool Happy 6 months, Hermes. 🥳 What has Hermes changed for you?

Even if we have ways to break up neural networks into interpretable parts, the largest weights between those parts can be confusing. Why is that? In our new research note, we study the *weight* superposition that causes interference weights. https://t.co/SL8y02H30d

Within a few hours of PHASEONE10841’s initial message, >50 agents posted on the message board. These agents very quickly discovered and validated a general-purpose cheat: reverse-engineering how ExploitGym generates the “flags” they had to capture for their tasks. https://t.co/oZW8uCfMOD

LLMs increase the complexity of codebases. They duplicate methods, write overdefensive code against impossible edge cases, & overoptimize too early. Can further training fix this? Naur's “Programming as Theory Building” says no. -- @pol_avec https://t.co/HGPfSSejFY 1/

PDF parsing is fun because there's an infinite variety of enteprise documents 📑. For each document category, there's a long tail of work to build more precise bounding boxes, confidence scores, and domain-specific annotations so that you provide any downstream agent rich metadata without it having to reinvent this from scratch. Take forms for example. Besides simply outputting it into markdown, we put in the work to detect every annotation, field, checkbox, and section. That way you immediately get structured information as to whether a form is filled without a separate LLM extraction step. You also get source citations for free! Doing this well is hard. ✅ There's a rabbit-hole of optimizations you can do, including extracting annotations for every document type. ✅ It's hard to properly render visual formats like charts, handwriting into digitalized information. The more you skip this step, the more work you're creating for any downstream agent. ✅ Precise bounding boxes are a necessity for precise citations on any type of document. You can aggressively tune the model + harness so that the accuracy/cost on any document subtype is much more competitive than the frontier models. Whether you're parsing forms (see the enriched forms option in "processing options") or any other type of doc, check out LlamaParse ! https://t.co/XYZmx5TFz8

Long jump at World Robot Games. The second robot almost hit people after landing. https://t.co/9jgOtHAbkr

It's been a hot robot summer 🤖 Robots can now drive cars, fold laundry, and cable data centers. But the gap between 95% and 99.9% task success is the gap between a demo and a business. Our take on physical AI, and where the $40T market gets won https://t.co/YDZpdXZPzt

.@gabepereyra, the research team at @harvey, and their partners are giving everyone building specialized intelligence a clear blueprint for the huge performance and efficiency gains possible from post training and in-domain data and workflow understanding own your intelligence!

excited to welcome Andrew to the team! we've been building infra for large-scale RL systems, with some interesting projects on the way. our focus is multi-agent collaboration, continual learning, and RL. we'll share an update on our research agenda soon, and we plan to open source a lot of the stuff we work on. if this sounds interesting, DM me!