Skip to main content

Custom Triton/CUDA kernels that complement Liger Kernel for LLM post-training

Project description

Chalk

Custom Triton/CUDA kernels that complement Liger Kernel.

pip install freesolo-chalk

Liger fuses the cross-entropy, activation, and RMSNorm paths. Chalk fills the gaps that matter for Freesolo's Flash post-training stack — fused GEMMs, the LoRA-delta matmuls, the QKV norm+RoPE epilogue, embedding gather, and FP8 frozen-base GEMMs — each behind a documented, benchmarked, opt-in entry point.

Chalk's repository layout and conventions intentionally mirror Liger Kernel.

Layout

src/chalk/
  ops/           # raw Triton/CUDA kernels + autograd.Function wrappers
  transformers/  # model-level installers that monkeypatch kernels into HF modules
  utils.py       # device detection helpers
test/            # correctness + gating tests (mirrors test/ops, test/transformers)

Design principles

  • Worker-side kernel library. Like Liger, chalk depends on torch + triton — it is meant to be installed where kernels actually run (the GPU worker), so consumers should depend on it from a gpu extra rather than their base install. Importing the top-level chalk package is still cheap (kernels lazy-load), so probing chalk.utils.infer_device() never forces a heavy import.
  • Complements, not replaces, Liger. Liger fuses CE / activation / RMSNorm; chalk fuses the GEMMs, LoRA delta, QKV epilogue, embedding, and FP8 base.
  • Safe fallback. Every installer is arch-gated, runs a numeric self-test on install, patches only frozen nn.Linear layers (never trainable / PEFT-wrapped layers), and silently falls back to the eager / Liger path on any import / compile / self-test failure.
  • Opt-in & evidence-based. Kernels are off unless explicitly enabled, and every kept kernel has end-to-end loss-curve evidence — not just a microbenchmark.

Development

pip install -e '.[dev]'
make checkstyle   # ruff check + format
make test         # pytest with coverage

Status

Intentionally minimal to start — kernels are landed one at a time under chalk/ops + chalk/transformers, each with correctness tests.

License

BSD-2-Clause. See LICENSE and NOTICE.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

freesolo_chalk-0.4.8.tar.gz (206.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

freesolo_chalk-0.4.8-py3-none-any.whl (113.8 kB view details)

Uploaded Python 3

File details

Details for the file freesolo_chalk-0.4.8.tar.gz.

File metadata

  • Download URL: freesolo_chalk-0.4.8.tar.gz
  • Upload date:
  • Size: 206.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.14 {"installer":{"name":"uv","version":"0.11.14","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for freesolo_chalk-0.4.8.tar.gz
Algorithm Hash digest
SHA256 3d1f8655507cc5e9b1c69efde2a1819291e8f8d6b4cfc5a034ddae00c64491cd
MD5 5adb0d30155b1e159e0028176707c972
BLAKE2b-256 3946ae821ff31e3e662025d19ad554259b1067d52d9f3490128100da49c82e40

See more details on using hashes here.

File details

Details for the file freesolo_chalk-0.4.8-py3-none-any.whl.

File metadata

  • Download URL: freesolo_chalk-0.4.8-py3-none-any.whl
  • Upload date:
  • Size: 113.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.11.14 {"installer":{"name":"uv","version":"0.11.14","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for freesolo_chalk-0.4.8-py3-none-any.whl
Algorithm Hash digest
SHA256 290a1ee5c3f385afefbfa441e3a4c87ccee3eeaee82e2527c1096641883e76d2
MD5 8584f70f7e6e962e6cafe15634dd0bf0
BLAKE2b-256 4e4b56ec6adce823da5e67f4c1bff56ea4a15b5b6089d6d5646a46bb7ac16edb

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page