ViperEkura 5244f1a8fc feat: add te-style scaled fp8 training via fp8_autocast ...

- per-tensor scales applied inside cublasLt via A_SCALE/B_SCALE
- delayed scaling: weight amax history ring, refresh every 16 steps
- quantize kernels emit atomic amax, device-side scale updates
- fp8_autocast context toggles aten::linear dispatch like torch.autocast
- fallback to bf16 when M/N not 16-aligned (fp8 gemm constraint)
- x/g scales delayed one step, reuse free atomic amax (no abs/max reduce)

2026-08-14 12:14:04 +08:00

..

kernels

feat: add te-style scaled fp8 training via fp8_autocast

2026-08-14 12:14:04 +08:00

tests

perf: flatten paged prefill tile dispatch

2026-08-09 23:12:53 +08:00

__init__.py

feat: add optional CUDA kernel system (csrc/) + fused GQA decode attention

2026-07-06 12:09:58 +08:00

CMakeLists.txt

feat: add fp8 training via cublasLt dispatch

2026-08-14 00:39:49 +08:00