CoChat
Sign Up
Cited by 1,814Topic ModelingAdvanced Neural Network ApplicationsDomain Adaptation and Few-Shot Learning

QLoRA: efficient finetuning of quantized LLMs

Open access
Dettmers, Tim, Pagnoni, Artidoro, Holtzman, Ari, Zettlemoyer, Luke·2023·Advances in Neural Information Processing Systems 36
DOI 10.52202/075280-0441
Open this paper in CoChat — free accountOpen PDF

Abstract

We present QLoRA, an efficient finetuning approach that reduces memory usage enough to finetune a 65B parameter model on a single 48GB GPU while preserving full 16-bit finetuning task performance. QLoRA backpropagates gradients through a frozen, 4-bit quantized pretrained language model into Low Rank Adapters~(LoRA). Our best model family, which we name Guanaco, outperforms all previous openly released models on the Vicuna benchmark, reaching 99.3% of the performance level of ChatGPT while only requiring 24 hours of finetuning on a single GPU. QLoRA introduces a number of innovations to save memory without sacrificing performance: (a) 4-bit NormalFloat (NF4), a new data type that is information theoretically optimal for normally distributed weights (b) double quantization to reduce the average memory footprint by quantizing the quantization constants, and (c) paged optimziers to manage memory spikes. We use QLoRA to finetune more than 1,000 models, providing a detailed analysis of instruction following and chatbot performance across 8 instruction datasets, multiple model types (LLaMA, T5), and model scales that would be infeasible to run with regular finetuning (e.g. 33B and 65B parameter models). Our results show that QLoRA finetuning on a small high-quality dataset leads to state-of-the-art results, even when using smaller models than the previous SoTA. We provide a detailed analysis of chatbot performance based on both human and GPT-4 evaluations showing that GPT-4 evaluations are a cheap and reasonable alternative to human evaluation. Furthermore, we find that current chatbot benchmarks are not trustworthy to accurately evaluate the performance levels of chatbots. A lemon-picked analysis demonstrates where Guanaco fails compared to ChatGPT. We release all of our models and code, including CUDA kernels for 4-bit training.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.52202/075280-0441

Cites

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • The first challenge arises from the massive scale of LLMs, which consist of billions or even trillions of parameters.

    “The first challenge arises from the massive scale of LLMs, which consist of billions or even trillions of parameters [287].”

    From A Survey of Large Language Models (2026) · cites this paper as [287]

    The Introduction describes the high cost of finetuning a 65-billion-parameter LLM, supporting the claim that LLMs can have billions of parameters and that their scale poses a challenge. The passages from our copy do not support the additional assertion that LLMs have trillions of parameters. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

No references could be resolved yet.

Research works printed but not linked: 67

  1. S. An, Y. Li, Z. Lin, Q. Liu, B. Chen, Q. Fu, W. Chen, N. Zheng, and J.-G. Lou. Input-tuning: Adapting unfamiliar inputs to frozen pretrained models. arXiv preprint arXiv:2203.03131, 2022. · arXiv
  2. A. Askell, Y. Bai, A. Chen, D. Drain, D. Ganguli, T. Henighan, A. Jones, N. Joseph, B. Mann, N. DasSarma, et al. A general language assistant as a laboratory for alignment. arXiv preprint arXiv:2112.00861, 2021. · arXiv
  3. S. H. Bach, V. Sanh, Z.-X. Yong, A. Webson, C. Raffel, N. V. Nayak, A. Sharma, T. Kim, M. S. Bari, T. Fevry, et al. Promptsource: An integrated development environment and repository for natural language prompts. arXiv preprint arXiv:2202.01279, 2022. · arXiv
  4. Y. Bai, A. Jones, K. Ndousse, A. Askell, A. Chen, N. DasSarma, D. Drain, S. Fort, D. Ganguli, T. Henighan, et al. Training a helpful and harmless assistant with reinforcement learning from human feedback. arXiv preprint arXiv:2204.05862, 2022. · arXiv
  5. Y. Bai, S. Kadavath, S. Kundu, A. Askell, J. Kernion, A. Jones, A. Chen, A. Goldie, A. Mirho- seini, C. McKinnon, et al. Constitutional ai: Harmlessness from ai feedback. arXiv preprint arXiv:2212.08073, 2022. · arXiv
  6. E. M. Bender, T. Gebru, A. McMillan-Major, and S. Shmitchell. On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency, pages 610-623, 2021.
  7. S. Biderman, H. Schoelkopf, Q. Anthony, H. Bradley, K. O'Brien, E. Hallahan, M. A. Khan, S. Purohit, U. S. Prashanth, E. Raff, et al. Pythia: A suite for analyzing large language models across training and scaling. arXiv preprint arXiv:2304.01373, 2023. · arXiv
  8. R. Bommasani, D. A. Hudson, E. Adeli, R. Altman, S. Arora, S. von Arx, M. S. Bernstein, J. Bohg, A. Bosselut, E. Brunskill, et al. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258, 2021. · arXiv
  9. T. Chen, B. Xu, C. Zhang, and C. Guestrin. Training deep nets with sublinear memory cost. arXiv preprint arXiv:1604.06174, 2016. · arXiv
  10. P. F. Christiano, J. Leike, T. Brown, M. Martic, S. Legg, and D. Amodei. Deep reinforcement learning from human preferences. Advances in neural information processing systems, 30, 2017.
  11. H. W. Chung, L. Hou, S. Longpre, B. Zoph, Y. Tay, W. Fedus, E. Li, X. Wang, M. De- hghani, S. Brahma, et al. Scaling instruction-finetuned language models. arXiv preprint arXiv:2210.11416, 2022. · arXiv
  12. T. Dettmers and L. Zettlemoyer. The case for 4-bit precision: k-bit inference scaling laws. arXiv preprint arXiv:2212.09720, 2022. · arXiv
  13. T. Dettmers, M. Lewis, Y. Belkada, and L. Zettlemoyer. LLM.int8(): 8-bit matrix multiplication for transformers at scale. Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, NeurIPS 2022, 2022.
  14. T. Dettmers, M. Lewis, S. Shleifer, and L. Zettlemoyer. 8-bit optimizers via block-wise quantization. 9th International Conference on Learning Representations, ICLR, 2022.
  15. A. E. Elo. The proposed uscf rating system. its development, theory, and applications. Chess Life, 22(8):242-247, 1967.
  16. E. Frantar, S. Ashkboos, T. Hoefler, and D. Alistarh. Gptq: Accurate post-training quantization for generative pre-trained transformers. arXiv preprint arXiv:2210.17323, 2022. · arXiv
  17. J. Fu, S.-K. Ng, Z. Jiang, and P. Liu. Gptscore: Evaluate as you desire. arXiv preprint arXiv:2302.04166, 2023. · arXiv
  18. A. Glaese, N. McAleese, M. Trębacz, J. Aslanides, V. Firoiu, T. Ewalds, M. Rauh, L. Weidinger, M. Chadwick, P. Thacker, et al. Improving alignment of dialogue agents via targeted human judgements. arXiv preprint arXiv:2209.14375, 2022. · arXiv
  19. S. Gururangan, S. Swayamdipta, O. Levy, R. Schwartz, S. R. Bowman, and N. A. Smith. Annotation artifacts in natural language inference data. arXiv preprint arXiv:1803.02324, 2018. · arXiv
  20. J. Henderson, S. Ruder, et al. Compacter: Efficient low-rank hypercomplex adapter layers. In Advances in Neural Information Processing Systems, 2021.
  21. D. Hendrycks, C. Burns, S. Basart, A. Zou, M. Mazeika, D. Song, and J. Steinhardt. Mea- suring massive multitask language understanding. In International Conference on Learning Representations, 2020.
  22. A. Holtzman, J. Buys, L. Du, M. Forbes, and Y. Choi. The curious case of neural text degeneration. In International Conference on Learning Representations, 2020.
  23. O. Honovich, T. Scialom, O. Levy, and T. Schick. Unnatural instructions: Tuning language models with (almost) no human labor. arXiv preprint arXiv:2212.09689, 2022. · arXiv
  24. N. Houlsby, A. Giurgiu, S. Jastrzebski, B. Morrone, Q. De Laroussilhe, A. Gesmundo, M. At- tariyan, and S. Gelly. Parameter-efficient transfer learning for nlp. In International Conference on Machine Learning, pages 2790-2799. PMLR, 2019.
  25. E. J. Hu, Y. Shen, P. Wallis, Z. Allen-Zhu, Y. Li, S. Wang, L. Wang, and W. Chen. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685, 2021. · arXiv
  26. S. Iyer, X. V. Lin, R. Pasunuru, T. Mihaylov, D. Simig, P. Yu, K. Shuster, T. Wang, Q. Liu, P. S. Koura, et al. Opt-iml: Scaling language model instruction meta learning through the lens of generalization. arXiv preprint arXiv:2212.12017, 2022. · arXiv
  27. A. Köksal, T. Schick, A. Korhonen, and H. Schütze. Longform: Optimizing instruction tuning for long text generation with corpus extraction. arXiv preprint arXiv:2304.08460, 2023. · arXiv
  28. A. Köpf, Y. Kilcher, D. von Rütte, S. Anagnostidis, Z.-R. Tam, K. Stevens, A. Barhoum, N. M. Duc, O. Stanley, R. Nagyfi, et al. Openassistant conversations-democratizing large language model alignment. arXiv preprint arXiv:2304.07327, 2023. · arXiv
  29. B. Lester, R. Al-Rfou, and N. Constant. The power of scale for parameter-efficient prompt tuning. arXiv preprint arXiv:2104.08691, 2021. · arXiv
  30. X. L. Li and P. Liang. Prefix-tuning: Optimizing continuous prompts for generation. arXiv preprint arXiv:2101.00190, 2021. · arXiv
  31. P. Liang, R. Bommasani, T. Lee, D. Tsipras, D. Soylu, M. Yasunaga, Y. Zhang, D. Narayanan, Y. Wu, A. Kumar, et al. Holistic evaluation of language models. arXiv preprint arXiv:2211.09110, 2022. · arXiv
  32. T. Liao, R. Taori, I. D. Raji, and L. Schmidt. Are we learning yet? a meta review of evaluation failures across machine learning. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2), 2021.
  33. H. Liu, D. Tam, M. Muqeeth, J. Mohta, T. Huang, M. Bansal, and C. A. Raffel. Few-shot parameter-efficient fine-tuning is better and cheaper than in-context learning. Advances in Neural Information Processing Systems, 35:1950-1965, 2022.
  34. Y. Liu, M. Ott, N. Goyal, J. Du, M. Joshi, D. Chen, O. Levy, M. Lewis, L. Zettlemoyer, and V. Stoyanov. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692, 2019. · arXiv
  35. S. Longpre, L. Hou, T. Vu, A. Webson, H. W. Chung, Y. Tay, D. Zhou, Q. V. Le, B. Zoph, J. Wei, et al. The flan collection: Designing data and methods for effective instruction tuning. arXiv preprint arXiv:2301.13688, 2023. · arXiv
  36. S. Min, M. Lewis, L. Zettlemoyer, and H. Hajishirzi. Metaicl: Learning to learn in context. arXiv preprint arXiv:2110.15943, 2021. · arXiv
  37. A. Nematzadeh, K. Burns, E. Grant, A. Gopnik, and T. Griffiths. Evaluating theory of mind in question answering. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2392-2400, 2018.
  38. OpenAI. Gpt-4 technical report. arXiv, 2023.
  39. L. Ouyang, J. Wu, X. Jiang, D. Almeida, C. Wainwright, P. Mishkin, C. Zhang, S. Agarwal, K. Slama, A. Ray, et al. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35:27730-27744, 2022.
  40. G. Park, B. Park, S. J. Kwon, B. Kim, Y. Lee, and D. Lee. nuqmm: Quantized matmul for efficient inference of large-scale generative language models. arXiv preprint arXiv:2206.09557, 2022. · arXiv
  41. B. Peng, C. Li, P. He, M. Galley, and J. Gao. Instruction tuning with gpt-4. arXiv preprint arXiv:2304.03277, 2023. · arXiv
  42. A. Poliak, J. Naradowsky, A. Haldar, R. Rudinger, and B. Van Durme. Hypothesis only baselines in natural language inference. In Proceedings of the Seventh Joint Conference on Lexical and Computational Semantics, pages 180-191, 2018.
  43. R. Pope, S. Douglas, A. Chowdhery, J. Devlin, J. Bradbury, A. Levskaya, J. Heek, K. Xiao, S. Agrawal, and J. Dean. Efficiently scaling transformer inference. arXiv preprint arXiv:2211.05102, 2022. · arXiv
  44. G. Qin and J. Eisner. Learning how to ask: Querying lms with mixtures of soft prompts. arXiv preprint arXiv:2104.06599, 2021. · arXiv
  45. V. Sanh, A. Webson, C. Raffel, S. H. Bach, L. Sutawika, Z. Alyafeai, A. Chaffin, A. Stiegler, T. L. Scao, A. Raja, et al. Multitask prompted training enables zero-shot task generalization. arXiv preprint arXiv:2110.08207, 2021. · arXiv
  46. M. Sap, R. LeBras, D. Fried, and Y. Choi. Neural theory-of-mind? on the limits of social intelligence in large lms. arXiv preprint arXiv:2210.13312, 2022. · arXiv
  47. T. L. Scao, A. Fan, C. Akiki, E. Pavlick, S. Ilić, D. Hesslow, R. Castagné, A. S. Luccioni, F. Yvon, M. Gallé, et al. Bloom: A 176b-parameter open-access multilingual language model. arXiv preprint arXiv:2211.05100, 2022. · arXiv
  48. S. Shaphiro and M. Wilk. An analysis of variance test for normality. Biometrika, 52(3):591-611, 1965.
  49. Y.-L. Sung, V. Nair, and C. A. Raffel. Training neural networks with fixed sparse masks. Advances in Neural Information Processing Systems, 34:24193-24205, 2021.
  50. R. Thoppilan, D. De Freitas, J. Hall, N. Shazeer, A. Kulshreshtha, H.-T. Cheng, A. Jin, T. Bos, L. Baker, Y. Du, et al. Lamda: Language models for dialog applications. arXiv preprint arXiv:2201.08239, 2022. · arXiv
  51. H. Touvron, T. Lavril, G. Izacard, X. Martinet, M.-A. Lachaux, T. Lacroix, B. Rozière, N. Goyal, E. Hambro, F. Azhar, et al. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971, 2023. · arXiv
  52. A. Wang, A. Singh, J. Michael, F. Hill, O. Levy, and S. R. Bowman. Glue: A multi- task benchmark and analysis platform for natural language understanding. arXiv preprint arXiv:1804.07461, 2018. · arXiv
  53. Y. Wang, Y. Kordi, S. Mishra, A. Liu, N. A. Smith, D. Khashabi, and H. Hajishirzi. Self-instruct: Aligning language model with self generated instructions. arXiv preprint arXiv:2212.10560, 2022. · arXiv
  54. Y. Wang, S. Mishra, P. Alipoormolabashi, Y. Kordi, A. Mirzaei, A. Arunkumar, A. Ashok, A. S. Dhanasekaran, A. Naik, D. Stap, et al. Super-naturalinstructions:generalization via declarative instructions on 1600+ tasks. In EMNLP, 2022.
  55. Y. Wang, S. Mishra, P. Alipoormolabashi, Y. Kordi, A. Mirzaei, A. Naik, A. Ashok, A. S. Dhanasekaran, A. Arunkumar, D. Stap, et al. Super-naturalinstructions: Generalization via declarative instructions on 1600+ nlp tasks. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 5085-5109, 2022.
  56. J. Wei, M. Bosma, V. Y. Zhao, K. Guu, A. W. Yu, B. Lester, N. Du, A. M. Dai, and Q. V. Le. Finetuned language models are zero-shot learners. arXiv preprint arXiv:2109.01652, 2021. · arXiv
  57. J. Wei, X. Wang, D. Schuurmans, M. Bosma, F. Xia, E. H. Chi, Q. V. Le, D. Zhou, et al. Chain-of-thought prompting elicits reasoning in large language models. In Advances in Neural Information Processing Systems, 2022.
  58. T. Wolf, L. Debut, V. Sanh, J. Chaumond, C. Delangue, A. Moi, P. Cistac, T. Rault, R. Louf, M. Funtowicz, et al. Huggingface's transformers: State-of-the-art natural language processing. arXiv preprint arXiv:1910.03771, 2019. · arXiv
  59. M. Wortsman, T. Dettmers, L. Zettlemoyer, A. Morcos, A. Farhadi, and L. Schmidt. Stable and low-precision training for large-scale vision-language models. arXiv preprint arXiv:2304.13013, 2023. · arXiv
  60. G. Xiao, J. Lin, M. Seznec, J. Demouth, and S. Han. Smoothquant: Accurate and efficient post-training quantization for large language models. arXiv preprint arXiv:2211.10438, 2022. · arXiv
  61. T. Xie, C. H. Wu, P. Shi, R. Zhong, T. Scholak, M. Yasunaga, C.-S. Wu, M. Zhong, P. Yin, S. I. Wang, et al. Unifiedskg: Unifying and multi-tasking structured knowledge grounding with text-to-text language models. arXiv preprint arXiv:2201.05966, 2022. · arXiv
  62. Z. Yang, P. Qi, S. Zhang, Y. Bengio, W. Cohen, R. Salakhutdinov, and C. D. Manning. Hotpotqa: A dataset for diverse, explainable multi-hop question answering. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2369-2380, 2018.
  63. Z. Yao, R. Y. Aminabadi, M. Zhang, X. Wu, C. Li, and Y. He. Zeroquant: Efficient and affordable post-training quantization for large-scale transformers. arXiv preprint arXiv:2206.01861, 2022. · arXiv
  64. E. B. Zaken, S. Ravfogel, and Y. Goldberg. Bitfit: Simple parameter-efficient fine-tuning for transformer-based masked language-models. arXiv preprint arXiv:2106.10199, 2021. · arXiv
  65. A. Zeng, X. Liu, Z. Du, Z. Wang, H. Lai, M. Ding, Z. Yang, Y. Xu, W. Zheng, X. Xia, et al. Glm-130b: An open bilingual pre-trained model. arXiv preprint arXiv:2210.02414, 2022. · arXiv
  66. S. Zhang, S. Roller, N. Goyal, M. Artetxe, M. Chen, S. Chen, C. Dewan, M. Diab, X. Li, X. V. Lin, et al. Opt: Open pre-trained transformer language models. arXiv preprint arXiv:2205.01068, 2022. · arXiv
  67. R. Zhong, K. Lee, Z. Zhang, and D. Klein. Adapting language models for zero-shot learning by meta-tuning on dataset and prompt collections. arXiv preprint arXiv:2104.04670, 2021. · arXiv

Cited by

Sorted by citations

Authors
Dettmers, T., Pagnoni, A., Holtzman, A., & Zettlemoyer, L.
ORCID
Venue
Advances in Neural Information Processing Systems 36
Publisher
Neural Information Processing Systems Foundation, Inc. (NeurIPS)
Published
2023
Pages
10088-10115
ISSN
1049-5258 (ISSN-L)
DOI
10.52202/075280-0441
OpenAlex
W7133233472
Cited by
1,814
References
0
Citations by year
2026: 985 · 2025: 414 · 2024: 338 · 2023: 76 · 2012: 1
FWCI
321.23
Type
conference paper
Language
en
Open access
Licence
cc-by
Locations
  • neural information processing systems (publishedVersion)
  • arXiv (Cornell University) (submittedVersion, cc-by)
  • arXiv (Cornell University) (cc-by)
Funders
University of Washington
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
Keywords
QLoRA, quantized large language models, efficient fine-tuning, low-rank adaptation, quantization, parameter-efficient fine-tuning

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review