Fort Lauderdale – October 21, 2026 -- Braid 1.9.5, a free local AI workspace for macOS, Windows and Linux, now works with vLLM, LM Studio and other OpenAI-compatible local inference servers in addition to Ollama.
Users choose how their models run
Developer Intersignal says the update lets people pick their own inference setup while staying inside Braid for conversations, sharing selected context between approved computers, and preparing publications. A clearer "Chat with Local AI" entry point helps new users get started on a single computer, and additional rigs can be connected as their needs grow. Streaming responses now display readable Markdown and code formatting.
"Bring your models. Braid brings the workspace," said David Seaman, founder of Intersignal.
Folio 0.4.0 turns original material into a public edition
The update includes Folio 0.4.0, a publishing studio that shows original material beside the public edition. It offers selective inclusion and redaction, local sensitive-detail warnings, and Editorial or Retro layouts.
Users review and explicitly approve their editions before exporting HTML or a website ZIP for hosting.
Local chat, network sharing and Folio cost nothing
Local chat, approved local network sharing and Folio are free in Braid 1.9.5.