# GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers

*Source* · `src-frantar-2022-gptq` · status: **n/a**

| field | value |
|---|---|
| kind | Peer-reviewed paper |
| authors | Elias Frantar, Saleh Ashkboos, Torsten Hoefler, Dan Alistarh |
| venue | ICLR 2023 |
| date | 2022-10-31 |
| url | https://arxiv.org/abs/2210.17323 |
| arxiv | 2210.17323 |
| accessed | 2026-10-24 |

## Notes

One-shot 3-4 bit weight quantization with approximate second-order information; 175B model quantized in about four GPU hours with minimal accuracy loss; about 3.25x (A100) and 4.5x (A6000) faster generation than FP16.

## Backlinks (computed)

**cited by:** [On-device small language models for characters](../technique/on-device-language-models.md)

## Citation

Elias Frantar, Saleh Ashkboos, Torsten Hoefler, Dan Alistarh (2022-10-31). *GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers*. ICLR 2023 https://arxiv.org/abs/2210.17323

---
Provenance: created in run `2026-10-24-techniques-llm-agents`.