commit 8ca405b15b8f64416e497ba68a00567963081170
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Fri Jul 10 13:50:23 2026 +0200

dev-python/comfy-kitchen-bin: drop 0.2.15

commit 2d4b1d366876f5a928c8139969fe991e67f5281d
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Fri Jul 10 13:49:52 2026 +0200

dev-python/comfy-kitchen-bin: add 0.2.18

commit d7bc56de3dd7a15917f475757a62d065db40eed2
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Wed Jul 1 09:17:39 2026 +0200

dev-python/comfy-kitchen-bin: drop 0.2.14

commit c122fa98fa14e2ba631e33037289579b2652b360
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Wed Jul 1 09:17:38 2026 +0200

dev-python/comfy-kitchen-bin: add 0.2.16

commit 83be69b6e7123c89cf5ccdf0537b3a297ec3a3f6
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Tue Jun 30 10:10:23 2026 +0200

dev-python/comfy-kitchen-bin: drop 0.2.13

commit fc4de505f84e20c191dffe243e43641af12e9fb7
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Tue Jun 30 10:10:22 2026 +0200

dev-python/comfy-kitchen-bin: add 0.2.15

commit 231ced06f2c7e5f1e82b377c6e79ab8d8034bc23
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sun Jun 28 14:12:48 2026 +0200

dev-python/comfy-kitchen-bin: drop 0.2.10, 0.2.12

commit c5191b8878ef2cb7a279b26def9541aed811a516
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sun Jun 28 14:12:48 2026 +0200

dev-python/comfy-kitchen-bin: add 0.2.14

commit 5046273815afd49da89ef6edc28cb7942674fef1
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Sat Jun 27 02:03:39 2026 +0200

dev-python/comfy-kitchen-bin: add 0.2.13

commit c3b547ccac181d6cd9c55fd1d8f1b947f53de2fd
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Fri Jun 26 13:28:00 2026 +0200

dev-python/comfy-kitchen-bin: add 0.2.12

commit 93f5abc78e61c9e3ac917f44b1da39f0d92600d7
Author: Ivan S. Titov <iohann.s.titov@gmail.com>
Date: Mon Jun 15 00:16:02 2026 +0200

dev-python/comfy-kitchen-bin: add 0.2.10

Fast diffusion-inference kernel library. Provides comfy_kitchen.apply_rope,
which ComfyUI's flux/lumina/z-image families call unconditionally through
comfy.quant_ops.ck during sampling -- so it is required for those models
despite the import being try/except-wrapped -- plus the FP8/FP4/NVFP4
quantization kernels.

Upstream ships no sdist. USE=cuda installs the cp312-abi3 x86_64 manylinux
wheel (CUDA + Triton kernels); without cuda the py3-none-any wheel (eager +
Triton backends, no CUDA) for the CPU/ROCm path.