Haru 4B
Haru-4B is a low-latency and multilingual full-duplex continuous spoken dialogue foundation model: it listens and speaks at the same time, the way people do on a phone call. It is a pretrained base model, released for research. The code is at : a PyTorch implementation with a web demo, an MLX implementation with a native macOS app, and a plugin for serving with vLLM-Omni.
Haru 4B launched on 12 November 2026. It ranked #17 on Daily Launch on 20 November 2026 with a Launch Score of 68, from trending score 311 on Hugging Face. It moved down 8 places from the day before.
What is Haru 4B?
Haru 4B is a model by higurashi-ai on Hugging Face. It is released under the Apache 2.0 license.
How to get Haru 4B
pip install -e ./haruFrom its model card on Hugging Face
More steps and the source are in its model card on Hugging Face.
Launch timeline
Launch Score, last 7 days
#16 of 226 in the roundup for the week of 16-22 November 2026.