Official PyTorch implementation for Extreme Compression of Large Language Models via Additive Quantization
Learn how to run the prequantized models using this Google Colab examples:
| Basic AQLM generation |
Streaming with GPU/CPU |
Inference with CUDA graphs (3x speedup) |
Fine-tuning with PEFT |
|---|---|---|---|
 |
|
|
|
This repository is currently designed to work with models of LLaMA, Mistral and Mixtral families.
The models reported below use full model fine-tuning as described in appendix A, with cross-entropy objective with teacher logits.
We provide a number of prequantized models:
| Model | AQLM scheme | WikiText 2 PPL | Model size, Gb | Hub link |
|---|---|---|---|---|
| Llama-2-7b | 1x16 | 5.92 | 2.4 | Link |
| Llama-2-7b | 2x8 | 6.69 | 2.2 | Link |
| Llama-2-7b | 8x8 | 6.61 | 2.2 | Link |
| Llama-2-13b | 1x16 | 5.22 | 4.1 | Link |
| Llama-2-70b | 1x16 | 3.83 | 18.8 | Link |
| Llama-2-70b | 2x8 | 4.21 | 18.2 | Link |
| Mixtral-8x7b | 1x16 | 3.35 | 12.6 | Link |
| Mixtral-8x7b-Instruct | 1x16 | - | 12.6 | Link |
AQLM quantization setpus vary mainly on the number of codebooks used as well as the codebook sizes in bits. The most popular setups, as well as inference kernels they support are:
| Kernel | Number of codebooks | Codebook size, bits | Scheme Notation | Accuracy | Speedup | Fast GPU inference | Fast CPU inference |
|---|---|---|---|---|---|---|---|
| Triton | K | N | KxN | - | Up to ~0.7x | ✅ | ❌ |
| CUDA | 1 | 16 | 1x16 | Best | Up to ~1.3x | ✅ | ❌ |
| CUDA | 2 | 8 | 2x8 | OK | Up to ~3.0x | ✅ | ❌ |
| Numba | K | 8 | Kx8 | Good | Up to ~4.0x | ❌ | ✅ |
To run the models, one would have to install an inference library:
pip install aqlm[gpu,cpu], specifying either gpu, cpu or both based on one's inference setting.
Then, one can use the familiar .from_pretrained method provided by the transformers library:
from transformers import AutoModelForCausalLM
quantized_model = AutoModelForCausalLM.from_pretrained(
"ISTA-DASLab/Llama-2-7b-AQLM-2Bit-1x16-hf",
trust_remote_code=True, torch_dtype="auto"
).cuda()Notice that torch_dtype should be set to either torch.float16 or "auto" on GPU and torch.float32 on CPU. After that, the model can be used exactly the same as one would use and unquantized model.
Install packages from requirements.txt:
pip install -r requirements.txtThe script will require downloading and caching locally the relevant tokenizer and the datasets. They will be saved in default Huggingface Datasets directory unless alternative location is provided by env variables. See relevant Datasets documentation section
When quantizing models with AQLM, we recommend that you use a subset of the original data the model was trained on.
For Llama-2 models, the closest available dataset is RedPajama . To load subset of RedPajama provide "pajama" in --dataset argument. This will process nsamples data and tokenize it using provided model tokenizer.
Additionally we provide tokenized Redpajama for LLama and Solar/Mistral models for 4096 context lengths stored in Hunggingface . To load it, use:
from huggingface_hub import hf_hub_download
hf_hub_download(repo_id="Vahe1994/AQLM", filename="data/name.pth",repo_type="dataset")To use downloaded data from HF, place it in data folder(optional) and set correct path to it in "--dataset" argument in main.py.
Warning: These subsets are already processed with the corresponding model tokenizer. If you want to quantize another model (e.g. mistral/mixtral), please re-tokenize the data with provided script in src/datautils.
One can optionally log the data to Weights and Biases service (wandb).
Run pip install wandb for W&B logging.
Specify $WANDB_ENTITY, $WANDB_PROJECT, $WANDB_NAME environment variables prior to running experiments. use --wandb argument to enable logging
This code was developed and tested using a several A100 GPU with 80GB GPU RAM.
You can use the --offload activations option to reduce VRAM usage.
For Language Model Evaluation Harness evaluation one needs to have enough memory to load whole model + activation tensors
on one or several devices.
AQLM quantization takes considerably longer to calibrate than simpler quantization methods such as GPTQ. This only impacts quantization time, not inference time.
For instance, quantizing a 7B model with default configuration takes about 1 day on a single A100 gpu. Similarly, quantizing a 70B model on a single GPU would take 10-14 days. If you have multiple GPUs with fast interconnect, you can run AQLM multi-gpu to speed up comparison - simply set CUDA_VISIBLE_DEVICES for multiple GPUs. Quantizing 7B model on two gpus reduces quantization time to ~14.5 hours. Similarly, quantizing a 70B model on 8 x A100 GPUs takes 3 days 18 hours.
If you need to speed up quantization without adding more GPUs, you may also increase --relative_mse_tolerance , --finetune_relative_mse_tolerance or set --init_max_points_per_centroid . However, that usually comes at a cost of reduced model accuracy.
The code requires the LLaMA model to be downloaded in Huggingface format and saved locally. The scripts below assume that $TRANSFORMERS_CACHE variable points to the Huggingface Transformers cache folder.
To download and cache the models, run this in the same environment:
from transformers import AutoTokenizer, AutoModelForCausalLM
model_name = "meta-llama/Llama-2-7b-hf" # or whatever else you wish to download
tokenizer = AutoTokenizer.from_pretrained(model_name, torch_dtype="auto")
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto")This script compresses the model and then tests its performance in terms of perplexity using WikiText2, C4, and Penn Treebank datasets.
The command to launch the script should look like this:
export CUDA_VISIBLE_DEVICES=0 # or e.g. 0,1,2,3
export MODEL_PATH=<PATH_TO_MODEL_ON_HUB>
export DATASET_PATH=<INSERT DATASET NAME OR PATH TO CUSTOM DATA>
export SAVE_PATH=/path/to/save/quantized/model/
export WANDB_PROJECT=MY_AQ_EXPS
export WANDB_NAME=COOL_EXP_NAME
python main.py $MODEL_PATH $DATASET_PATH --nsamples=1024 \
--num_codebooks=1 --nbits_per_codebook=16 --in_group_size=8 \
--relative_mse_tolerance=0.01 --finetune_relative_mse_tolerance=0.001 \
--finetune_batch_size=32 --local_batch_size=1 --offload_activations \
--wandb --save $SAVE_PATH
Main CLI arguments:
CUDA_VISIBLE_DEVICES- by default, the code will use all available GPUs. If you want to use specific GPUs (or one GPU), use this variable.MODEL_PATH- a path to either hugginface hub (e.g. meta-llama/Llama-2-7b-hf) or a local folder with transformers model and a tokenizer.DATASET_PATH- either a path to calibration data (see above) or a standard dataset[c4, ptb, wikitext2]- for llama-2 models, you can use
DATASET_PATH=./data/red_pajama_n=1024_4096_context_length.pthfor a slice of RedPajama (up to 1024 samples)
- for llama-2 models, you can use
--nsamples- the number of calibration data sequences. If this parameter is not set, take all calibration data avaialble.--num_codebooks- number of codebooks per layer--nbits_per_codebook- each codebook will contain 2 ** nbits_per_codebook vectors--in_group_size- how many weights are quantized together (aka "g" in the arXiv paper)--finetune_batch_size- (for fine-tuning only) the total number of sequences used for each optimization step--local_batch_size- when accumulating finetune_batch_size, process this many samples per GPU per forward pass (affects GPU RAM usage)--relative_mse_tolerance- (for initial calibration) - stop training when (current_epoch_mse / previous_epoch_mse) > (1 - relative_mse_tolerance)--finetune_relative_mse_tolerance- same, but for fine-tuning--offload_activations-- during calibration, move activations from GPU memory to RAM. This reduces VRAM usage while slowing calibration by ~10% (depending on your hardware).--save-- path to save/load quantized model. (see also:--load)--wandb- if this parameter is set, the code will log results to wandb
There are additional hyperparameters aviailable. Run python main.py --help for more details on command line arguments, including compression parameters.
To perform zero-shot evaluation, we use Language Model Evaluation Harness framework with slight modifications. This repository contains a copy of LM Evaluation Harness repo from early 2023 in lm-eval-harness folder.
Before running the code make sure that you have all the requirements and dependencies of lm-eval-harness installed. To install them run:
pip install -r lm-evaluation-harness/requirements.txt
The main script launching the evaluation procedure is lmeval.py .
export CUDA_VISIBLE_DEVICES=0,1,2,3 # optional: select GPUs
export QUANTZED_MODEL=<PATH_TO_SAVED_QUANTIZED_MODEL_FROM_MAIN.py>
export MODEL_PATH=<INSERT_PATH_TO_ORIINAL_MODEL_ON_HUB>
export DATASET=<INSERT DATASET NAME OR PATH TO CUSTOM DATA>
export WANDB_PROJECT=MY_AQ_LM_EVAL
export WANDB_NAME=COOL_EVAL_NAME
python lmeval.py \
--model hf-causal \
--model_args pretrained=$MODEL_PATH,dtype=float16,use_accelerate=True \
--load $QUANTZED_MODEL \
--tasks winogrande,piqa,hellaswag,arc_easy,arc_challenge \
--batch_size 1To convert a model into a Hugging Face compatible format, use convert_to_hf.py with corresponding arguments:
--model- the original pretrained model (corresponds toMODEL_PATHofmain.py, e.g.meta-llama/Llama-2-7b-hf).--in_path- the folder containing an initially quantized model (corresponds to--saveofmain.py).--out_path- the folder to savetransformersmodel to.
The conversion automatically
If you want to contribute something substantial (more than a typo), please open an issue first.
We use black and isort for all pull requests. Before committing your code run black . && isort .
If you found this work useful, please consider citing:
@misc{egiazarian2024extreme,
title={Extreme Compression of Large Language Models via Additive Quantization},
author={Vage Egiazarian and Andrei Panferov and Denis Kuznedelev and Elias Frantar and Artem Babenko and Dan Alistarh},
year={2024},
eprint={2401.06118},
archivePrefix={arXiv},
primaryClass={cs.LG}
}