HeartMuLa-RL-oss-3B-20260123 (Burn Format)

This repository contains Burn-format weights for the upstream model:

The published artifacts are packaged as Burn Pack (.bpk) archives. The repository also includes Rust tooling to regenerate the raw export manifest and Burn pack from the upstream checkpoint.

Repository Contents

Published model assets at the repository root:

  • heartmula.bpk
  • tokenizer.json
  • gen_config.json
  • config.json
  • LICENSE

Conversion and packaging sources included in the repository:

  • convert.sh: one-step export-and-pack script that writes heartmula.bpk at the repository root
  • src/bin/export_heartmula_raw.rs: exports the upstream safetensors checkpoint into .npy tensors plus a manifest file
  • src/main.rs: packs exported tensors into Burn Pack archives
  • Cargo.toml, Cargo.lock: Rust dependencies for the packer

Rebuilding the Artifacts

Prerequisites

  • A Rust toolchain capable of building the packer in this repository
  • Local copies of the upstream checkpoints and support files arranged like this:
CKPT_ROOT/
β”œβ”€β”€ HeartMuLa-RL-oss-3B-20260123/
β”œβ”€β”€ tokenizer.json
└── gen_config.json

This conversion repo ships the Apache-2.0 LICENSE directly and copies it into the output artifact directory during conversion.

One-Step Conversion

./convert.sh CKPT_ROOT

This regenerates heartmula.bpk at the repository root and removes the temporary artifact directory when finished.

Downloads last month
9
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for maolandaw/HeartMuLa-RL-oss-3B-20260123

Finetuned
(2)
this model