Weightsmithregistry

Model registry

Weights are artefacts. Treat them like it.

Content-addressed storage, resumable multi-hundred-gigabyte transfers, deduplication across fine-tunes, and a signed provenance record that travels with the file.

 content addressed resumable signed provenance
$ ws pull acme/atlas-70b@v3 --dest ./weights
resolving acme/atlas-70b@v3 → sha256:8f21c0a2…
142 GB · 38 shards · 12 already local (dedup)
pulling 130 GB at 3.8 Gbit/s … 4m41s
verified · provenance signature ok
128,400
model versions
4.2PB
unique bytes stored
11.8PB
logical size before dedup
2.9PB
pulled last month

Built for files that do not fit in a git repo

Chunk-level dedup

A LoRA on top of a 70B base stores the delta, not the base. Across our registry that is a 2.8× saving and a much faster pull.

Transfers that survive

Resumable, parallel, verified per chunk. A 400 GB pull that dies at 92% resumes at 92%.

Provenance that verifies

Training data manifest, base model, hyperparameters and evaluation results, signed by the publisher's key.

Licences that travel

SPDX identifiers and use restrictions in the manifest, surfaced by the CLI before a pull, not buried in a card.

Safe formats only

safetensors and GGUF. Pickle-based checkpoints are rejected at push because they execute code on load.

Mirrored

Four regions plus a torrent swarm for the largest public releases. Pulls pick the fastest source.

Deduplication

Why the registry is smaller than its contents

Most published models are derivatives. Chunking at a content boundary rather than a file boundary means shared tensors are stored once, regardless of how the shards were split.

  • 11.8 PB logical, 4.2 PB physical
  • A typical fine-tune adds 0.4–3% new bytes
  • Pulls reuse chunks already on your disk from other models
  • Verification is per chunk, so corruption is local and cheap to fix
1base1sft1dpo0lora-a1lora-b1quant-40quant-81merge1110

Common questions

LFS stores whole files and re-uploads a shard when a single tensor changes. At 140 GB per version that is the difference between four minutes and four hours.