In this video I walk through the full setup of Hermes Agent from scratch. We plug it into Gemma 4 running locally through Ollama and set up self-hosted Firecrawl for private web search. By the end, you have a fully local, fully private AI agent connected to Telegram with no paid APIs required.
At its core, Muse Spark is a natively multimodal reasoning model. Unlike previous iterations that "stitched" vision and text together, Muse Spark was rebuilt from the ground up to integrate visual information across its internal logic. This architectural shift enables "visual chain of thought," allowing the model to annotate dynamic environments—identifying the components of a complex espresso machine or correcting a user's yoga form via side-by-side video analysis.
Z.ai, also known as Zhupai AI, a Chinese AI startup best known for its powerful, open source GLM family of models, has unveiled GLM-5.1 today under a permissive MIT License, allowing for enterprises to download, customize and use it for commercial purposes.
The new GLM-5.1 is designed to work autonomously for up to eight hours on a single task, marking a definitive shift from vibe coding to agentic engineering.
In this video, I share the inside scoop on how to build agent-readable skills that actually compound:
• Why the description field is where most skills go to die
• How agent-first design changes handoffs and contracts
• What three-tier skill architecture looks like for teams
• Where community repositories fill the domain-specific gap
Karpathy proposes something simpler and more loosely, messily elegant than the typical enterprise solution of a vector database and RAG pipeline.
Instead, he outlines a system where the LLM itself acts as a full-time "research librarian"—actively compiling, linting, and interlinking Markdown (.md) files, the most LLM-friendly and compact data format.
By diverting a significant portion of his "token throughput" into the manipulation of structured knowledge rather than boilerplate code, Karpathy has surfaced a blueprint for the next phase of the "Second Brain"—one that is self-healing, auditable, and entirely human-readable.
Gemma 4 arrives as four distinct models organized into two deployment tiers. The "workstation" tier includes a 31B-parameter dense model and a 26B A4B Mixture-of-Experts model — both supporting text and image input with 256K-token context windows. The "edge" tier consists of the E2B and E4B, compact models designed for phones, embedded devices, and laptops, supporting text, image, and audio with 128K-token context windows.
Microsoft AI, the tech giant’s research lab, announced the release of three foundational AI models on Thursday that can generate text, voice, and images.
The release signals Microsoft’s continued push to build out its own stack of multimodal AI models — and compete with rival AI labs — even though it remains tied to OpenAI.