llmcompressor
is an easy-to-use library for optimizing models for deployment with vllm
, including:
- Comprehensive set of quantization algorithms for weight-only and activation quantization
- Seamless integration with Hugging Face models and repositories
safetensors
-based file format compatible withvllm
- Large model support via
accelerate
✨ Read the announcement blog here! ✨
Big updates have landed in LLM Compressor! Check out these exciting new features:
- Axolotl Sparse Finetuning Integration: Easily finetune sparse LLMs through our seamless integration with Axolotl. Learn more here.
- AutoAWQ Integration: Perform low-bit weight-only quantization efficiently using AutoAWQ, now part of LLM Compressor. Note: This integration should be considered experimental for now. Enhanced support, including for MoE models and improved handling of larger models via layer sequential pipelining, is planned for upcoming releases. See the details.
- Day 0 Llama 4 Support: Meta utilized LLM Compressor to create the FP8-quantized Llama-4-Maverick-17B-128E, optimized for vLLM inference using compressed-tensors format.
- Activation Quantization: W8A8 (int8 and fp8)
- Mixed Precision: W4A16, W8A16
- 2:4 Semi-structured and Unstructured Sparsity
- Simple PTQ
- GPTQ
- AWQ
- SmoothQuant
- SparseGPT
Please refer to docs/schemes.md for detailed information about available optimization schemes and their use cases.
pip install llmcompressor
Applying quantization with llmcompressor
:
- Activation quantization to
int8
- Activation quantization to
fp8
- Weight only quantization to
int4
using GPTQ - Weight only quantization to
int4
using AWQ - Quantizing MoE LLMs
- Quantizing Vision-Language Models
- Quantizing Audio-Language Models
Deep dives into advanced usage of llmcompressor
:
Let's quantize TinyLlama
with 8 bit weights and activations using the GPTQ
and SmoothQuant
algorithms.
Note that the model can be swapped for a local or remote HF-compatible checkpoint and the recipe
may be changed to target different quantization algorithms or formats.
Quantization is applied by selecting an algorithm and calling the oneshot
API.
from llmcompressor.modifiers.smoothquant import SmoothQuantModifier
from llmcompressor.modifiers.quantization import GPTQModifier
from llmcompressor import oneshot
# Select quantization algorithm. In this case, we:
# * apply SmoothQuant to make the activations easier to quantize
# * quantize the weights to int8 with GPTQ (static per channel)
# * quantize the activations to int8 (dynamic per token)
recipe = [
SmoothQuantModifier(smoothing_strength=0.8),
GPTQModifier(scheme="W8A8", targets="Linear", ignore=["lm_head"]),
]
# Apply quantization using the built in open_platypus dataset.
# * See examples for demos showing how to pass a custom calibration set
oneshot(
model="TinyLlama/TinyLlama-1.1B-Chat-v1.0",
dataset="open_platypus",
recipe=recipe,
output_dir="TinyLlama-1.1B-Chat-v1.0-INT8",
max_seq_length=2048,
num_calibration_samples=512,
)
The checkpoints created by llmcompressor
can be loaded and run in vllm
:
Install:
pip install vllm
Run:
from vllm import LLM
model = LLM("TinyLlama-1.1B-Chat-v1.0-INT8")
output = model.generate("My name is")
- If you have any questions or requests open an issue and we will add an example or documentation.
- We appreciate contributions to the code, examples, integrations, and documentation as well as bug reports and feature requests! Learn how here.