Aillex is a fully-local talking AI assistant — her voice, brain, memory and animated 3D avatar all run on a single consumer GPU. No cloud, no subscription, no data leaving your machine. We build her in the open and publish every step so you can build your own. Start with the architecture overview, watch the first episode — The Voice You’re Hearing Right Now? I Built It — follow Building Aillex — the whole story in order, or find her on YouTube. A new DIY AI Brief lands every Monday.
We benched LTX-2.5 for two full rounds and got confident, wrong answers, because the official templates ship an LLM prompt rewriter that is on by default and buried inside a subgraph. Here is how to find it, how to actually turn it off, and what the model does once your prompts reach it intact.
The bridge nobody documents: connect a local Ollama brain to a real-time 3D VRM face with voice, lip-sync and expressions. The four services, the ports they talk on, and the two traps that cost us days — the thinking-mode stall and the open-mic mistake.
Our nightly pipeline writes, renders and publishes a video at 4:30am with nobody watching. Then it published one it shouldn’t have. The four-layer gate we built afterward — deterministic block, vision judge, fail-closed, and the undo button you build before you need it.
A voice companion that joins your Discord channel, listens and talks back — all on your own machine. Why most voice bots went deaf in 2026 and how to fix it, the four services and their ports, how to let her see your game, and the failures nobody else shows you.
A local agent becomes genuinely useful the day you can teach it: skills are small text files it reads before acting. Here are the two we taught ours this week — summarize any YouTube link, answer only from my own notes — both downloadable, plus the GPU tax nobody warns you about and the honest bill.
A local model can join your Minecraft world and actually play — gather, craft, follow you, even recover from its own mistakes. Here is the working setup, the two silent failures that will waste your evening if nobody warns you, and an honest account of what an 8B model can and cannot do in there.
The same edit three ways with a 4-step Lightning LoRA, Flux Kontext and full Qwen. What each trade-off actually costs.
One node takes three reference images. Person plus product, person plus scene, and the group shot - with measured timings.
One generation, seventeen outputs: the flat image, every object on transparency, and a free background plate.
We ran the same edit with and without a mask. The no-mask version won. Here’s when a mask helps and when it costs you.