Hugging Face seeks MXFP4 backward kernels for GPT-OSS training
Forward-only 4-bit support forces a costly upcast, blocking native low-precision fine-tuning on modern NVIDIA GPUs.
By tensor#triton
PyTorch adds Triton JIT cache hydrate from verified bundles
A precompile import path can stage a checked runtime-cache bundle so GPU kernel launches skip first-use compile and autotune.
By tensortorch.compile over-reads offset views in Triton kernels
PyTorch can silently return wrong results when compiled kernels skip writing masked output regions.
By tensor