Memory-optimized SongGeneration (v2 Large) for 16GB VRAM GPUs. Features 8-bit µ-law KV-caching, fused layers, and SDPA/Triton integration.
linux generative-audio pytorch nvidia triton llama music-generation quantization rocm sdpa nvidia-gpu amd-gpu mu-law wsl2 text-to-music generative-ai audio-ai songgeneration
-
Updated
Apr 13, 2026 - Python