Hugging Face and SkyPilot jointly upgrade Hub to SkyPilot's store: hf first-class storage backend, mounting any model, dataset, or Space repo directly to GPU tasks on 20+ clouds (AWS, GCP, Lambda, CoreWeave, etc.) and Kubernetes/Slurm clusters via hf:// URLs, with the core implemented at two ends: (1) the new hf-mount FUSE backend — each read() only pulls the few bytes needed, avoiding full replication, so GPUs are "almost immediately working", with local on-disk cache keeping repeat reads hitting local; (2) Xet-based content-defined chunking (CDC) — splitting model weights into ~64 KB chunks, re-uploading only the truly modified parts, uniformly exposed as store: hf in SkyPilot's file_mounts. Benchmarks use the same qwen-sft.yaml switching --infra across three clouds to run Qwen3.5-4B SFT, model first read about 30 seconds (peak ~500 MB/s), 8.43 GB checkpoint written to bucket hits ~168 MB/s on AWS L40S, ~123 MB/s on GCP L4, ~112 MB/s on Lambda H100. Xet pushes dedup down to Parquet row-level append — internal tests appending 10K rows to a 100K-row table upload only ~10 MB, re-uploading an already-stored 8.43 GB blob only takes about 8 seconds. This change is the first time "data lives on Hub, compute runs anywhere" has been made a product-grade contract, breaking the long-standing deployment constraint that "wherever the object storage is, GPU scheduling is pinned there" — the data-movement cost of cross-cloud training and inference is compressed to near zero.