Skip to main content
Cadenza runs the same compute — residual RL, the GRD LoRA+RL loop, distillation, LoRA fine-tuning, frozen-VLA inference — across very different hardware. Rather than make you pass device flags, the CLI ships a single acceleration layer that detects your hardware once and applies the right knobs transparently. Nobody calls it directly; every training and inference path runs through it.

​
Targets

The layer chooses device, thread/interop counts, float32 matmul precision (TF32/bf16), bf16 autocast, torch.compile, channels_last, and int8 for deploy — and applies each technique only where it has surface.

​
Safety first

Every aggressive step is guarded: if anything fails, it falls back to eager fp32 so acceleration can never break or slow a run. The layer is also empirical about traps — for example, bf16 on Apple-Silicon CPU is emulated (≈100× slower), so it is never blindly assumed; the layer probes before committing.

​
Kill-switches

Two environment variables make it debuggable: Commands that train or profile print a one-line acceleration summary (e.g. in env residual init and env residual bench), so you can see exactly what was applied on your box.
Torch is imported lazily — the core CLI stays light, and only the rl / lora / vla paths that already need torch trigger the acceleration layer. Hardware detection uses the Python standard library only.