m-a-p releases YuE2-3B, an open music generation model trending on HF

m-a-p · hf · 2026-09-10

m-a-p's YuE2-3B is trending on Hugging Face: an open text-to-audio music generation model supporting Chinese and English, featuring agentic editing capabilities and backed by an arXiv paper (2503.08638).

Related event: Open-source music model YuE2-3B launches with editable symbolic composition, claims to rival Suno v5(10 posts)→

Original post →

More from Multimodal

• Tripo Launches First World-Building 3D Hackathon With $80K+ Prizes — majidmanzarpour · 2026-09-11
• Dev trains a diffusion model on optical flow maps from video — pixlpa · 2026-09-11
• SpreeAI cuts virtual try-on inference time 1.7x and halves peak GPU memory with Lambda — TheZachMueller · 2026-09-11
• Suno Accused of Charging Users to License Their Likeness for Model Training — NickPassig · 2026-09-11
• Free Open-Source Cinematic LoRA Takes MiniMax H3 From 1K to 4K in ComfyUI — jugernaut126 · 2026-09-11
• Longtime Suno user ready to cancel over blocked song downloads — NickPassig · 2026-09-11

Multimodal channel →