higurashi-ai/Haru-4B
audio-to-audio · ♥ 864 · ↓ 8.6K
- Taskaudio-to-audio
- Downloads7323
- Likes4230
- Licenseapache-2.0
audio-to-audio · ♥ 864 · ↓ 8.6K
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.
Excerpt from the model card on Hugging Face ↗.
higurashi-ai/Haru-4B was first seen on Hugging Face · Trending models on Nov 14, 2026. Since then it has appeared in 195 checks on 9 different days. Its best position was #1 of 30, reached on Nov 18, 12:44 UTC. It is #7 right now, holding steady at #7 over the last 6 checks.
Position on Hugging Face · Trending models at every check (lower is better).
Best rank per day (UTC).
🔥 as reported by Hugging Face.
Archived daily, weekly and monthly lists that include this entry.
Yes. As of Nov 22, 16:45 UTC it is #7, holding steady at #7 over the last 6 checks.
#1 on Hugging Face · Trending models, on Nov 18, 12:44 UTC. Trendoday checks this list every 60 minutes.
It was first seen on Nov 14, 2026 and has been on the list on 9 days (195 checks) since.
It is tagged audio-to-audio on Hugging Face and has 7323 downloads under the apache-2.0 license.