llm-layer-prefetch/pyproject.toml at main · SergiuDeveloper/llm-layer-prefetch · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
[build-system]
requires = ["hatchling"]
build-backend = "hatchling.build"

[project]
name = "llm-layer-prefetch"
version = "1.0.0"
description = "Stream transformer blocks layer-by-layer from disk to GPU with a pipelined prefetch queue (disk → CPU RAM → pinned RAM → GPU)."
readme = "README.md"
license = { text = "MIT" }
requires-python = ">=3.11"
keywords = ["llm", "inference", "streaming", "offloading", "pytorch", "cuda", "safetensors"]
classifiers = [
    "Development Status :: 4 - Beta",
    "Intended Audience :: Developers",
    "Intended Audience :: Science/Research",
    "License :: OSI Approved :: MIT License",
    "Programming Language :: Python :: 3",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Programming Language :: Python :: 3.13",
    "Programming Language :: Python :: 3.14",
    "Topic :: Scientific/Engineering :: Artificial Intelligence",
]
dependencies = [
    "tqdm>=4.0",
    "torch",
]

[project.optional-dependencies]
examples = [
    "transformers>=4.40",
    "huggingface_hub>=0.23",
]

[tool.hatch.build.targets.wheel]
packages = ["src/layer_streamer"]