On October 19‑20 2026 I posted a series of X updates that announced a really fast open‑source HTML artifact builder powered by libfx, a “mini v0” demo built on the same library, and new AI models (Mercury 3 and MiniMax M3.1 Flash) on the Vercel AI Gateway with special pricing.
We open sourced a really fast HTML artifact builder. It's powered by libfx, making the agent durable.
Building the low-level plumbing for v0 to survive server crashes, flaky LLMs, function limits was a huge pain.
Now it's easy, free, and OSS: https://t.co/aUP1vKeIsP.
The mini v0 demo on libfx showed how to design and prototype in seconds, exposing 200+ models from the AI Gateway and emitting pure HTML artifacts. The demo repo is available at https://t.co/0LvclJqJLH.
I announced Mercury 3 with a 50 % discount and a 1 M‑token context window, plus MiniMax M3.1 Flash free through Oct 22, to showcase high‑throughput diffusion and agentic task performance.
Community replies praised the speed (“Congrats!”, “so cool!”, “🫡”) and highlighted a Barber‑shop landing page built in less than a minute as a benchmark.
These updates build on earlier announcements in Part 5 and fit within the broader AI agent ecosystem.