Text Generation
Transformers
Safetensors
English
Chinese
Russian
yue2
music-generation
orbitquant
quantization
4-bit precision
custom-code
8-bit precision
Instructions to use WaveCut/YuE2-3B-OrbitQuant-W4A4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use WaveCut/YuE2-3B-OrbitQuant-W4A4 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="WaveCut/YuE2-3B-OrbitQuant-W4A4")# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("WaveCut/YuE2-3B-OrbitQuant-W4A4", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use WaveCut/YuE2-3B-OrbitQuant-W4A4 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "WaveCut/YuE2-3B-OrbitQuant-W4A4" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "WaveCut/YuE2-3B-OrbitQuant-W4A4", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/WaveCut/YuE2-3B-OrbitQuant-W4A4
- SGLang
How to use WaveCut/YuE2-3B-OrbitQuant-W4A4 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "WaveCut/YuE2-3B-OrbitQuant-W4A4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "WaveCut/YuE2-3B-OrbitQuant-W4A4", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "WaveCut/YuE2-3B-OrbitQuant-W4A4" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "WaveCut/YuE2-3B-OrbitQuant-W4A4", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use WaveCut/YuE2-3B-OrbitQuant-W4A4 with Docker Model Runner:
docker model run hf.co/WaveCut/YuE2-3B-OrbitQuant-W4A4
File size: 6,446 Bytes
3ea3da7 | 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 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 | import logging
import os
from shutil import which, move
import subprocess
import sys
from pathlib import Path
from setuptools import Extension, find_packages, setup
from setuptools.command.build import build
from setuptools.command.build_ext import build_ext
logger = logging.getLogger(__name__)
def get_backend() -> str:
"""Detect the backend by inspecting torch."""
import torch
if torch.version.cuda is not None:
return "cuda"
elif torch.version.hip is not None:
return "rocm"
elif torch.backends.mps.is_available():
return "metal"
elif hasattr(torch.version, "xpu") and torch.version.xpu is not None:
return "xpu"
else:
return "cpu"
def is_sccache_available() -> bool:
return which("sccache") is not None
def is_ccache_available() -> bool:
return which("ccache") is not None
def is_ninja_available() -> bool:
return which("ninja") is not None
def _make_cmake_args(cfg: str) -> tuple[list[str], list[str]]:
"""Build CMake and build arguments from the current environment."""
cmake_generator = os.environ.get("CMAKE_GENERATOR", "")
cmake_args = [
f"-DPython3_EXECUTABLE={sys.executable}",
f"-DCMAKE_BUILD_TYPE={cfg}", # not used on MSVC, but no harm
]
build_args: list[str] = []
if "CMAKE_ARGS" in os.environ:
cmake_args += [item for item in os.environ["CMAKE_ARGS"].split(" ") if item]
if not cmake_generator or cmake_generator == "Ninja":
try:
import ninja
ninja_executable_path = Path(ninja.BIN_DIR) / "ninja"
cmake_args += [
"-GNinja",
f"-DCMAKE_MAKE_PROGRAM:FILEPATH={ninja_executable_path}",
]
except ImportError:
pass
if is_sccache_available():
cmake_args += [
"-DCMAKE_C_COMPILER_LAUNCHER=sccache",
"-DCMAKE_CXX_COMPILER_LAUNCHER=sccache",
"-DCMAKE_CUDA_COMPILER_LAUNCHER=sccache",
"-DCMAKE_HIP_COMPILER_LAUNCHER=sccache",
"-DCMAKE_OBJC_COMPILER_LAUNCHER=sccache",
"-DCMAKE_OBJCXX_COMPILER_LAUNCHER=sccache",
]
elif is_ccache_available():
cmake_args += [
"-DCMAKE_C_COMPILER_LAUNCHER=ccache",
"-DCMAKE_CXX_COMPILER_LAUNCHER=ccache",
"-DCMAKE_CUDA_COMPILER_LAUNCHER=ccache",
"-DCMAKE_HIP_COMPILER_LAUNCHER=ccache",
"-DCMAKE_OBJC_COMPILER_LAUNCHER=ccache",
"-DCMAKE_OBJCXX_COMPILER_LAUNCHER=ccache",
]
num_jobs = os.getenv("MAX_JOBS", None)
if num_jobs is not None:
num_jobs = int(num_jobs)
logger.info("Using MAX_JOBS=%d as the number of jobs.", num_jobs)
else:
try:
# os.sched_getaffinity() isn't universally available, so fall
# back to os.cpu_count() if we get an error here.
num_jobs = len(os.sched_getaffinity(0))
except AttributeError:
num_jobs = os.cpu_count()
nvcc_threads = os.getenv("NVCC_THREADS", None)
if nvcc_threads is not None:
nvcc_threads = int(nvcc_threads)
logger.info(
"Using NVCC_THREADS=%d as the number of nvcc threads.", nvcc_threads
)
num_jobs = max(1, num_jobs // nvcc_threads)
cmake_args += ["-DNVCC_THREADS={}".format(nvcc_threads)]
build_args += [f"-j{num_jobs}"]
if sys.platform == "win32":
build_args += ["--config", cfg]
return cmake_args, build_args
class CMakeExtension(Extension):
def __init__(self, name: str, sourcedir: str = "") -> None:
super().__init__(name, sources=[], py_limited_api=True)
self.sourcedir = os.fspath(Path(sourcedir).resolve())
class CMakeBuild(build_ext):
def build_extension(self, ext: CMakeExtension) -> None:
ext_fullpath = Path.cwd() / self.get_ext_fullpath(ext.name)
extdir = ext_fullpath.parent.resolve()
debug = int(os.environ.get("DEBUG", 0)) if self.debug is None else self.debug
cfg = "Debug" if debug else "Release"
cmake_args, build_args = _make_cmake_args(cfg)
cmake_args = [f"-DCMAKE_LIBRARY_OUTPUT_DIRECTORY={extdir}{os.sep}"] + cmake_args
build_temp = Path(self.build_temp) / ext.name
if not build_temp.exists():
build_temp.mkdir(parents=True)
subprocess.run(
["cmake", "-S", ext.sourcedir, "-B", str(build_temp), *cmake_args],
cwd=build_temp,
check=True,
)
subprocess.run(
["cmake", "--build", str(build_temp), *build_args], cwd=build_temp, check=True
)
if sys.platform == "win32":
# Move the dylib one folder up for discovery.
for filename in os.listdir(extdir / cfg):
move(extdir / cfg / filename, extdir / filename)
class BuildKernel(build):
"""Custom command to build and locally install the kernel."""
description = "Build the kernel and install via the local_install CMake target"
user_options = []
def initialize_options(self) -> None:
super().initialize_options()
def finalize_options(self) -> None:
super().finalize_options()
def run(self) -> None:
project_root = Path(__file__).parent
debug = int(os.environ.get("DEBUG", 0))
cfg = "Debug" if debug else "Release"
cmake_args, build_args = _make_cmake_args(cfg)
build_temp = project_root / "_cmake_build"
build_temp.mkdir(parents=True, exist_ok=True)
subprocess.run(
["cmake", "-S", str(project_root), "-B", str(build_temp), *cmake_args],
cwd=project_root,
check=True,
)
subprocess.run(
["cmake", "--build", str(build_temp), "--target", "local_install", *build_args],
cwd=project_root,
check=True,
)
backend = get_backend()
ops_name = f"_orbitquant_gemv_{backend}_2r2jqnkz5xfei"
setup(
name="orbitquant_gemv",
# The version is just a stub, it's not used by the final build artefact.
version="0.1.0",
ext_modules=[CMakeExtension(f"orbitquant_gemv.{ops_name}")],
cmdclass={"build_ext": CMakeBuild, "build_kernel": BuildKernel},
packages=find_packages(where="torch-ext", include=["orbitquant_gemv*"]),
package_dir={"": "torch-ext"},
zip_safe=False,
install_requires=["torch"],
python_requires=">=3.9",
) |