stabilityai
/

stable-audio-3-optimized

audio-generation

Model card Files Files and versions

stable-audio-3-optimized

114 GB

Ctrl+K

Ctrl+K

5 contributors

History: 80 commits

cortexelus's picture

sm_120 sa3-m: add the fp8 medium-DiT tier (176 fp8 GEMM + 96 bf16 fused MHA, 1.43-1.95x vs fp16mixed)

af5680d verified 6 days ago

MLX
Add MLX base-model DiT weights for LoRA training (#2) 22 days ago
onnx
sa3-m fp8: calibrated dit_fp8.onnx (real-conditioning activation + per-channel weight scales, @ryanontheinside #47; refs unchanged dit_fp8lin.onnx.data) 6 days ago
tensorRT
sm_120 sa3-m: add the fp8 medium-DiT tier (176 fp8 GEMM + 96 bf16 fused MHA, 1.43-1.95x vs fp16mixed) 6 days ago
tflite
tflite: replace naive same-l/enc_w8a8-dyn.tflite with GPTQ (torch-free, real-audio calibrated) 29 days ago
.gitattributes

1.78 kB
Add ONNX files for per-arch TRT compilation 3 months ago
LICENSE.md

11.9 kB
Add LICENSE.md 3 months ago
LICENSE_GEMMA.md

10.5 kB
Add LICENSE_GEMMA.md 3 months ago
NOTICE

97 Bytes
Add NOTICE 3 months ago
README.md

5.21 kB
Update model card 3 months ago
Stable_Audio_3.0_Thumbnail_1x1.png

1.46 MB
xet

Add logo 3 months ago