# Newsletter 19 November 2026: NVIDIA releases 4B draft model for Qwen3.8-27B

[← All newsletters](/en/nieuwsbrieven)

# Newsletter Thursday 19 November 2026

By Ivo Donker · 19 November 2026 — compiled with AI assistance
Daily update on the AI tools we track. Headlines come from public sources (X, GitHub, Hacker News); the Dutch commentary and summary are machine‑written. Each headline links to our own news page on that topic; the source link is shown beside it.

## In brief

NVIDIA has published a new draft model with 4 billion parameters on Hugging Face. This model is designed for speculative decoding. Indie developers can use it to increase the testing speed of the Qwen3.8‑27B model.

## Repos, releases & updates

- [NVIDIA Releases a 4B Draft Model to Accelerate Qwen3.8-27B NVIDIA has released a new 4B speculative decoding draft model on Hugging Face, the](/nieuws-repos-releases#i-9db3b023e2) · [source: X](https://x.com/Oluwaphilemon1/status/2123202361506029602) · 2026-11-18NVIDIA has placed a 4‑billion‑parameter draft model on Hugging Face, which can help indie developers test Qwen3.8‑27B models faster.

[← 17 November 2026](/en/nieuwsbrief-2026-11-17) · [Archive](/en/nieuwsbrieven)
