quantprobe/pyproject.toml at master · FedericoTs/quantprobe · 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
38
39
40
41
42
43
44
45
46
47
48
49
[build-system]
build-backend = "setuptools.build_meta"
requires = [ "setuptools>=68" ]

[project]
name = "quantprobe"
version = "1.37.0"
description = """\
  Probe-then-quantize for LLMs: measure a model's fragility curve, plan bit/tier placement by the tiered decode law, \
  and emit ready-to-run llama.cpp recipes.\
  """
readme = "README.md"
keywords = [ "gguf", "inference", "llama.cpp", "llm", "moe", "quantization" ]
license = { text = "MIT" }
authors = [ { name = "Federico Sciuca" } ]
requires-python = ">=3.11"
classifiers = [
  "License :: OSI Approved :: MIT License",
  "Programming Language :: Python :: 3 :: Only",
  "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 = [ "gguf>=0.9", "requests>=2.28" ]
urls.Homepage = "https://github.com/FedericoTs/quantprobe"
scripts.quantprobe = "quantprobe.cli:main"

[dependency-groups]
dev = [
  "bandit==1.9.4",
  "pre-commit==4.6.2",
  "pyproject-fmt==2.28.0",
  "ruff==0.16.4",
]

[tool.setuptools]
packages = [ "quantprobe" ]
package-data.quantprobe = [ "recipes/*.json" ]

[tool.ruff]
line-length = 100
# BLE001/S110/S112: this tool probes external CLIs (ollama, nvidia-smi, llama-bench);
# swallowing their failures is intentional, not an accident.
lint.ignore = [ "BLE001", "S110", "S112" ]

[tool.pyproject-fmt]
keep_full_version = true