commit eea247964c4fa6120a3827c45ae24e4f0fdf2274
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Wed Jul 8 10:52:52 2026 +0200

dev-python/humming-kernels: add 0.1.10

commit 48b449bc24da2a668fee7bcc0c0ccd34df7aa3f6
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Fri Jun 26 20:35:40 2026 +0200

dev-python/humming-kernels: drop 0.1.5, 0.1.6

commit 4b11e489a8d2fc1f1ff8a9db456c957e6b2df8a6
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Fri Jun 26 19:59:53 2026 +0200

dev-python/humming-kernels: add 0.1.7

commit c1c264f251147591f29f444cd71d9adddc6ca607
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sat Jun 20 10:09:53 2026 +0200

dev-python/humming-kernels: add 0.1.6

commit e62192514368500fd30be63f7c13d6a45331b8ad
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Tue Jun 16 14:28:03 2026 +0200

dev-python/humming-kernels: add 0.1.5

Prebump ahead of its consumer: vllm's latest release (0.23.0) still pins
humming-kernels[cu13]==0.1.4, so nothing pulls 0.1.5 yet. Kept current
for when a future vllm bumps the cu13 pin; the existing ~0.1.4 pin in
vllm-0.23.0[humming] is unaffected.

commit 1fc4e0437a00dcaf4e556a74d8ee6c40ed6b905a
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sun Jun 14 23:40:12 2026 +0200

dev-python/humming-kernels: new package

humming-kernels provides the `humming` Python namespace that vLLM's
quantization registry imports unconditionally on CUDA builds
(quantization/__init__.py pulls in humming.py, whose external import is
gated only by current_platform.is_cuda() with no fallback). Without it,
loading any quantized model under vllm[cuda] aborts with
ModuleNotFoundError: No module named 'humming', regardless of which
quantization method was actually requested.

JIT GEMM kernel library: a pure-Python wheel that compiles its bundled
CUDA sources at runtime via the system nvcc. 0.1.2 and 0.1.4 match the
humming-kernels pins in vllm 0.22.1 and 0.23.0 requirements/cuda.txt.