audio model · stable-audio · Linux
Can I run Stable Audio Open 1.0 on Nvidia DGX Spark (128GB)?
Yes. Stable Audio Open 1.0 runs on Nvidia DGX Spark (128GB) at fp32 (~15 GB of ~112 GB usable).
Runs at fp32 using ~15 GB of ~112 GB usable.
- Peak memory
- ~15 GB
- Usable on device
- ~112 GB
- Device memory
- 128 GB
- Quant
- fp32
How to run it
Use stable-audio-tools or diffusers at fp32. It loads as a single model and runs on a GPU.
- Type
- Music and audio generation
- Parameters
- 1.3B
- Peak memory
- ~15 GB at fp32
- License
- Stability Community License
- Memory
- 128 GB unified
- Usable for weights
- ~112 GB
- Power draw
- ~140 W
- Best runtime
- llama.cpp (CUDA) / Ollama (CUDA)
You could also run
Run Stable Audio Open 1.0 on other hardware
FAQ
Can Nvidia DGX Spark (128GB) run Stable Audio Open 1.0?
Yes. Stable Audio Open 1.0 runs on Nvidia DGX Spark (128GB) at fp32 (~15 GB of ~112 GB usable).
How much memory does Stable Audio Open 1.0 need?
Nvidia DGX Spark (128GB) has room to spare. At fp32 the realistic peak is ~15 GB of memory.
What do I use to run Stable Audio Open 1.0 locally?
Stable Audio Open 1.0 runs in stable-audio-tools or diffusers (among others). It needs a GPU.
Sources
-
arxiv.org · 1 source
-
docs.nvidia.com · 2 sources
-
github.com · 1 source
-
huggingface.co · 2 sources
-
nvidia.com · 1 source
-
nvidianews.nvidia.com · 1 source
-
ollama.com · 1 source
-
tomshardware.com · 1 source
VRAM figures are sourced peak-usage anchors at the noted quant; catalog updated 2026-11-01. See methodology.