YouTube Videos - 2026

Your LLM no longer reads every token. The newest open models from Alibaba, Moonshot, DeepSeek, and Z.AI have replaced full attention in most or all of…

This video is sponsored by Crusoe. If you have used the OpenAI fine-tuning API, you already know this one.

Giving an AI agent database access has an obvious problem: experimenting means DDL.

I upgraded my local RAG chatbot to a 35B Qwen mixture-of-experts at 32K context, running entirely on a MacBook.

KV-cache quantization in MLX: I benchmarked TurboQuant on Apple Silicon to find out how much memory does it really save, and where it breaks.

This is the third video I've made with @replit . The first two were single-feature demos: Design Mode for web, Expo for mobile.

In this video, I close a real open issue in cache-commander, my open-source Rust project, in one session — using the new Cursor and Composer 2.

🚨 Get 12% off any Token Plan tier: https://platform.minimax.io/subscribe/coding-plan?code=9PuYDYrfpN&source=link Resources: - M2.7 model card…

Three frontier open-weight MoE reasoning models — Trinity Large Thinking (~400B), DeepSeek R1 (671B), and Kimi K2 Thinking (1T) — are compared side by…

Today I'm happy to ship something I've been building for a while: Canopy, a native macOS app for running hyper-productive parallel Claude Code sessions…

Three frontier open-weight MoE reasoning models — Trinity Large Thinking (~400B), DeepSeek R1 (671B), and Kimi K2 Thinking (1T) — are compared side by…

Two years after parts 1 (https://youtu.be/kw7S-3s50uk) and 2 (https://youtu.be/fXBBwCIA0Ds), the quantization landscape has changed completely.

No API keys. No cloud. No per-token cost. Just your Mac. In this video, I show you how to run Arcee AI's Trinity Mini (26B parameters, 3B active) locally…

Bojan Jakimovski, an ML engineer, took Arcee AI's open-source Trinity Mini model and turned it into a biomedical specialist — extracting drug-protein…

I built a code review app powered by @ArceeAI 's new Trinity Large model and deployed it to a @HuggingFace Space.

In this video, we use @replit to build a Kanban Task Board, a project management app where you can organize work across boards using drag-and-drop cards.

In this silent demo, watch as I use Claude Code's team orchestration to spawn 5 specialized AI agents that review my personal website (julien.org) in…

I built a code review app powered by Arcee AI's new Trinity Large model and deployed it to a Hugging Face Space.

In this tutorial, we build a fully functional iOS app from scratch in just a few minutes using Replit, Expo, and React Native.

Can AI coding tools actually handle parallel tasks without breaking everything? I put Verdent to the test.

In this video, we build a fully local RAG chatbot that runs entirely on a MacBook - no cloud APIs, no usage costs, complete privacy.