Adding GPT-NeoX and Pythia support for GPTQ quantization/4bit inference and beyond.

I have uploaded neox.py and neox2.py, my two different approaches to adding neox support; if anyone wishes to contribute to this project based off of GPTQ-for-LLaMA and assist, please do - Each script loads and goes past datasets, then I'm snagged on how to properly reference layers in NeoX. Any advice, code change PRs, suggestions would help a long way. Also, all three AutoTokenizer.from_pretrained(model, use_fast=False) in datautils.py might need to be changed to AutoTokenizer.from_pretrained(model, use_fast=True), as NeoX only supports fast tokenization; not sure if it needs its own entry in datautils.py.

Let's get GPT-NeoX and Pythia support for 4bit GPTQ!

GPTQ-for-LLaMa

4 bits quantization of LLaMa using GPTQ

GPTQ is SOTA one-shot weight quantization method

This code is based on GPTQ

Result

Model(LLaMa-7B)	Bits	group-size	Wikitext2	PTB	C4
FP16	16	-	5.67	8.79	7.05
RTN	4	-	6.28	9.68	7.70
GPTQ	4	-	6.79	10.67	8.28
GPTQ	4	64	6.16	9.66	7.52
RTN	3	-	25.66	61.25	28.19
GPTQ	3	-	20.86	37.54	22.19
GPTQ	3	64	12.24	16.77	9.55

Model(LLaMa-13B)	Bits	group-size	Wikitext2	PTB	C4
FP16	16	-	5.08	8.06	6.58
RTN	4	-	5.52	8.62	6.96
GPTQ	4	-	5.35	8.40	6.82
GPTQ	4	64	5.18	8.18	6.66
RTN	3	-	11.41	21.21	13.20
GPTQ	3	-	6.80	10.45	8.31
GPTQ	3	64	5.50	8.60	7.00

Quantizing the model requires a large amount of CPU memory. For example, quantizing a LLaMa-13b model requires 42gb, and LLaMa-33b requires more memory than 64gb.

Depending on the GPUs/drivers, there may be a difference in performance, which decreases as the model size increases.(IST-DASLab/gptq#1)

According to GPTQ paper, As the size of the model increases, the difference in performance between FP16 and GPTQ decreases.

Installation

If you don't have conda, install it first.

conda create --name gptq python=3.9 -y
conda activate gptq
conda install pytorch torchvision torchaudio pytorch-cuda=11.6 -c pytorch -c nvidia
# Or, if you're having trouble with conda, use pip with python3.9:
# pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116

git clone https://github.com/qwopqwop200/GPTQ-for-LLaMa
cd GPTQ-for-LLaMa
pip install -r requirements.txt

Dependencies

torch: tested on v1.13.1+cu116
transformers: tested on v4.27.0.dev0(required)
datasets: tested on v2.10.1
safetensors: tested on v0.3.0
(to run 4-bit kernels: setup for compiling PyTorch CUDA extensions, see also https://pytorch.org/tutorials/advanced/cpp_extension.html, tested on CUDA 11.6)

All experiments were run on a single NVIDIA RTX3090.

Language Generation

LLaMa

# Compute full precision (FP16) results
CUDA_VISIBLE_DEVICES=0 python llama.py decapoda-research/llama-7b-hf c4
# Run RTN baseline and compute results
CUDA_VISIBLE_DEVICES=0 python llama.py decapoda-research/llama-7b-hf c4 --wbits 4 --nearest
# Run GPTQ and compute results
CUDA_VISIBLE_DEVICES=0 python llama.py decapoda-research/llama-7b-hf c4 --wbits 4 --groupsize 64

To run other LLaMa models replace llama-7b-hf with one of: llama-13b-hf, llama-30b-hf, llama-65b-hf.

ZeroShot

See zeroShot/ folder.

CUDA Kernels

# Install kernels
python setup_cuda.py install

# Benchmark performance for FC2 layer of LLaMa-7B
CUDA_VISIBLE_DEVICES=0 python test_kernel.py

# Benchmark language generation with 4-bit LLaMa-7B:

# Save compressed model
CUDA_VISIBLE_DEVICES=0 python llama.py decapoda-research/llama-7b-hf c4 --wbits 4 --save llama7b-4bit.pt
# Or save compressed `.safetensors` model
CUDA_VISIBLE_DEVICES=0 python llama.py decapoda-research/llama-7b-hf c4 --wbits 4 --save_safetensors llama7b-4bit.safetensors
# Benchmark generating a 2048 token sequence with the saved model
CUDA_VISIBLE_DEVICES=0 python llama.py decapoda-research/llama-7b-hf c4 --wbits 4 --load llama7b-4bit.pt --benchmark 2048 --check
# Benchmark FP16 baseline, note that the model will be split across all listed GPUs
CUDA_VISIBLE_DEVICES=0,1,2,3,4 python llama.py decapoda-research/llama-7b-hf c4 --benchmark 2048 --check

# model inference with the saved model
CUDA_VISIBLE_DEVICES=0 python llama_inference.py decapoda-research/llama-7b-hf --wbits 4 --load llama7b-4bit.pt --text "this is llama"
# model inference with the saved model with offload(This is very slow. This is a simple implementation and could be improved with technologies like flexgen(https://github.com/FMInference/FlexGen).
CUDA_VISIBLE_DEVICES=0 python llama_inference_offload.py decapoda-research/llama-7b-hf --wbits 4 --load llama7b-4bit.pt --text "this is llama" --pre_layer 16
It takes about 180 seconds to generate 45 tokens(5->50 tokens) on single RTX3090 based on LLaMa-65B. pre_layer is set to 50.

CUDA Kernels support 2,3,4,8 bits.

Basically, 4-bit quantization is recommended.

cuda kernel does not support group size.

Memory Usage

Model	Bits	memory(MiB)	benchmark(ppl)	Wikitext2	PTB	C4	checkpoint size(GB)
LLaMa-7B with FP16	16	13940	5.23	5.67	8.79	7.05	12.5
LLaMa-13B with FP16	16	OOM	-	5.08	8.06	6.58	24.2
LLaMa-7B with GPTQ	8	7748	5.39	5.67	8.81	7.08	6.5
LLaMa-13B with GPTQ	8	14570	5.00	5.09	8.06	6.61	12.4
LLaMa-7B with GPTQ	4	4740	6.23	6.79	10.67	8.28	3.5
LLaMa-13B with GPTQ	4	8410	5.14	5.35	8.40	6.82	6.5
LLaMa-33B with GPTQ	4	19499	4.59	4.45	7.58	6.22	16.9
LLaMa-7B with GPTQ	3	3852	11.43	17.94	31.44	19.65	2.75
LLaMa-13B with GPTQ	3	6870	5.58	6.77	10.29	8.34	5.06
LLaMa-33B with GPTQ	3	15499	5.10	5.78	8.98	7.38	12.94
LLaMa-7B with GPTQ	2	3076	4152	30749	45936	5045	2.0
LLaMa-13B with GPTQ	2	5275	6903	13203	1384	8.34	5.06
LLaMa-33B with GPTQ	2	11635	66.58	-	-	-	8.9

Acknowledgements

This code is based on GPTQ

Thanks to Meta AI for releasing LLaMa, a powerful LLM.

Digitous / GPTQ-for-GPT-NeoX