Instructions to use WaveCut/YuE2-3B-OrbitQuant-W4A4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

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
```

YuE2-3B-OrbitQuant-W4A4 / src /kernel-source /orbitquant-native /build.toml

WaveCut

Update packed inference to OrbitQuant0.9.6 and native1.0.3

f0c91ec verified 4 days ago

Raw

History Blame Contribute Delete

1.83 kB

	[general]
	name = "orbitquant-packed-matmul"
	version = 1
	edition = 5
	license = "Apache-2.0"
	backends = ["cpu", "cuda", "metal"]
	upstream = "https://github.com/iamwavecut/OrbitQuant"
	source = "https://huggingface.co/WaveCut/orbitquant-packed-matmul"

	[general.hub]
	repo-id = "WaveCut/orbitquant-packed-matmul"

	[torch]
	src = [
	"torch-ext/torch_binding.cpp",
	"torch-ext/torch_binding.h",
	]

	[torch.stable-abi]
	cpu = "2.11"

	[kernel.packed_matmul_cpu]
	backend = "cpu"
	depends = ["torch"]
	include = ["orbitquant_packed_matmul_cpu"]
	src = [
	"orbitquant_packed_matmul_cpu/cpu_isa.cpp",
	"orbitquant_packed_matmul_cpu/cpu_kernel_args.h",
	"orbitquant_packed_matmul_cpu/cpu_pool.cpp",
	"orbitquant_packed_matmul_cpu/cpu_pool.h",
	"orbitquant_packed_matmul_cpu/cpu_threads.cpp",
	"orbitquant_packed_matmul_cpu/cpu_threads.h",
	"orbitquant_packed_matmul_cpu/packed_adaln_cpu.cpp",
	"orbitquant_packed_matmul_cpu/packed_matmul_cpu.cpp",
	"orbitquant_packed_matmul_cpu/packed_matmul_cpu.h",
	"orbitquant_packed_matmul_cpu/packed_matmul_scalar.cpp",
	"orbitquant_packed_matmul_cpu/packed_matmul_neon.cpp",
	"orbitquant_packed_matmul_cpu/packed_matmul_x86_avx512.cpp",
	"orbitquant_packed_matmul_cpu/quantize_activations_cpu.cpp",
	]

	[kernel.packed_matmul_cpu_x86_avx2]
	backend = "cpu"
	depends = ["torch"]
	include = ["orbitquant_packed_matmul_cpu"]
	cxx-flags = ["$<$<CXX_COMPILER_ID:MSVC>:/arch:AVX2>"]
	src = [
	"orbitquant_packed_matmul_cpu/cpu_msvc_avx2.cpp",
	"orbitquant_packed_matmul_cpu/packed_matmul_x86.cpp",
	]

	[kernel.packed_matmul_cuda]
	backend = "cuda"
	depends = ["torch"]
	src = ["orbitquant_packed_matmul_cuda/packed_matmul.cu"]

	[kernel.packed_matmul_metal]
	backend = "metal"
	depends = ["torch"]
	src = [
	"orbitquant_packed_matmul_metal/packed_matmul.mm",
	"orbitquant_packed_matmul_metal/packed_matmul.metal",
	]