YuE2-3B open music model lands on Hugging Face, tops WildSongBench over Suno v5/v6
AdinaYakup · x · 2026-09-14
YuE2-3B from the M-A-P community is now on Hugging Face (cc-by-nc-4.0, Chinese and English). On 192 WildSongBench prompts it scores 6.9632 (best-of-8), beating Suno v5 (6.8721), Suno v6 (6.5562) and Suno v6 Wild (6.4195) across all open and proprietary models. Features: lyrics+style to full song, editable ABC scores for melody/chords, agent-based editing, and unquantized local inference on a 24GB GPU. Built on an AR–NAR Mixture-of-Transformers backbone with flow matching and a VAE for 48kHz stereo output.
Related event: Open-Source Music Model YuE2 Beats Suno v5/v6 on WildSongBench(4 posts)→
More from Multimodal
- Gemini 4 Carbon vs Opus 5.5 hands-on: 3x cheaper at $4.10 and twice as fast — vitaliychiley · 2026-10-31
- Surreal AI fairytale: a town obsessed with turning tears into yarn — gizakdag · 2026-10-31
- ComfyUI extension adds one-click saving of preview images with metadata — sitefall · 2026-10-31
- DeepMind scientist directs MKUltra microdrama with Gemini agents for ~$35 total — korymath · 2026-10-31
- Using a gamepad-controlled Godot camera path to drive Wan video generation — droidsculptor · 2026-10-31
- Customuse launches Decals to fix 3D AI models' blurry logos and text — SimplyAnnisa · 2026-10-31