Version control for
model weights

Hexz brings versioned, deduplicated storage for ML checkpoints to your terminal. Open source and self-hosted.

pip install hexz

$ hexz

Initialize repos, commit checkpoints, diff snapshots, and mount weights for serving — all from one binary.

hexz

Create a weight repository backed by your object store.

$ hexz init s3://ml-checkpoints/llm-7b

Initialized empty weight repository
  remote: s3://ml-checkpoints/llm-7b
  config: .hexz/config.toml

Built for checkpoint infrastructure

Content-addressed storage

Weights are chunked and addressed by hash. Identical data is stored once regardless of how many snapshots reference it.

Thin snapshots

Each commit records references to chunks, not copies. New checkpoints only write the bytes that actually changed.

Random access

Read individual tensors or slices without downloading full checkpoint files. Useful for serving and evaluation.

Your object store

Backends plug into S3, GCS, or compatible APIs. No proprietary service or mandatory control plane.

Reproducible lineage

Tag, diff, and inspect any snapshot. Know exactly which weights backed a deployment or evaluation run.

Single binary

No runtime dependencies, no daemon, no cluster. Download the binary and start versioning weights.

Try hexz on the command line

Early access is limited. Leave your email and we'll share progress, invite a small group of teams to shape the roadmap, and let you know when the CLI is ready.