Higurashi AI releases Haru-4B, a voice model that listens and speaks at once

Nov 13, 2026
The open research model holds full-duplex spoken conversations in several languages and runs on PyTorch, MLX and vLLM-Omni.

Higurashi AI has released Haru-4B on Hugging Face, a multilingual spoken dialogue model that listens and speaks at the same time, as people do on a phone call, its model card says. It is a pretrained base model released for research.

The code includes a PyTorch version with a web demo, an MLX version with a native macOS app and a plugin for serving with vLLM-Omni. Users give it a character card and a voice. Higurashi warns that it is not a reliable source of facts, does not follow instructions like an assistant and can imitate a reference voice, which should only be used with permission.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Daily AI Brief.

Also, consider following us on social media:

AI Brief AI Brief (X)
Free newsletter

Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates

Guide
AI meeting notetakers for workplace conversations

AI meeting notetakers for workplace conversations

A guide to capture methods, meeting notes and team workflows

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.