Open-weight AI modelsSat 31 Oct 2026 · 5 of 15

MonsterHouse: AI-Powered Grumpy Pirate House for Halloween

Verified · 10/10 supported1 cited · 1 min read

MonsterHouse is an interactive projection created for Halloween that listens to trick-or-treaters and replies in the voice of a grumpy old pirate ship, softening when visitors say “please” . The system runs on a single workstation and uses NVIDIA Nemotron Voice Agent components: Nemotron ASR for speech recognition, Nemotron 3.5 Bolt E4B 4 GB as the language model, Magpie TTS Zeroshot for voice synthesis, and Audio2Face‑3D for mouth animation in Unreal Engine 5 . A LoRA adapter applied to the Bolt E4B model provides the grumpy pirate personality for the replies . The creator designed MonsterHouse to let the garage door deliver treats while being more grumpy than scary, offering a fun, unique interaction for Halloween visitors .

Why it matters. Trick-or-treaters can now engage with a talking house that delivers treats while responding to manners, enhancing the Halloween experience for children .

Earlier on Open-weight AI models

Fri 30 Oct 2026Granite Switch 4.1.1 patch adds multilingual ASR and faster Decision adapterGranite Switch 4.1.1 is a patch release that is fully backwards compatible with version 4.1 and requires no migration. The update adds an updated Granite Speech ASR model that provides multilingual speech‑to‑text in 10 languages with native punctuation and capitalization, ready for downstream use. It also introduces a new Decision adapter in granitelib‑core that delivers improved performance on the Jev Decision Index test set. The base model remains unchanged, with only adapters updated, enabling rapid and efficient releases. Hugging Face says it is developing tutorials and an Adapter SDK so developers and third‑party providers can build, test, and ship their own adapters like software updates.Tue 27 Oct 2026Growth of Chinese Open-Weight AI Model Use Raises Security Scrutinycohere.com reports that companies and government contractors adopt open-weight AI models believing that running them on their own servers keeps data secure. cohere.com reports that security officials in Ottawa, Washington DC, London, Berlin and Tokyo see two options: closed-API services or self-hosted open-weight models. cohere.com reports that thousands of engineering teams have chosen to self-host open-weight models, increasingly using Chinese models. cohere.com reports that Chinese models accounted for 41 % of downloads on Hugging Face in the year before March 2026, and handled more than 30 % of AI work on OpenRouter each week from February through early July 2026. cohere.com reports that U.S. lawmakers opened an investigation into corporate use of Chinese AI models in spring 2026.Fri 23 Oct 2026Hugging Face launches OlmoEarth v1.3 with global embeddings and web explorerHugging Face released OlmoEarth v1.3, a state-of-the-art satellite embedding model. The model provides precomputed annual embeddings for the entire planet at 10‑meter resolution in both 64‑ and 128‑dimensional formats. A new browser‑based OlmoEarth Embeddings Explorer lets users explore and customize maps without sign‑in or advanced technical skills. OlmoEarth v1.3 reads a year of Sentinel‑1 radar and Sentinel‑2/Landsat multispectral time series to produce a 128‑dimensional vector for each 10‑meter pixel, with the first 64 dimensions usable independently. This embedding approach replaces the previous fine‑tuning workflow that required GPU access and imagery download, enabling cheap, reusable features for many tasks.

Next · Health researchFDA grants accelerated approval for first plasmid DNA immunotherapy, Voceryq, for adult RRP →