# BitDelta: Your Fine-Tune May Only Be Worth One Bit

arXiv: https://arxiv.org/abs/2402.10193 · HTML: https://mlsanity.com/paper/2402.10193/

## Key facts

- Cited by 39 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 62 references that are in the map.
- First posted on arXiv 15 Feb 2024.

## Builds on (most cited first, 30 of 40)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [QLoRA: Efficient Finetuning of Quantized LLMs](https://mlsanity.com/paper/2305.14314/) (2023)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time](https://mlsanity.com/paper/2203.05482/) (2022)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot](https://mlsanity.com/paper/2301.00774/) (2023)
- [Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation](https://mlsanity.com/paper/2108.12409/) (2021)
- [Accelerating Large Language Model Decoding with Speculative Sampling](https://mlsanity.com/paper/2302.01318/) (2023)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads](https://mlsanity.com/paper/2401.10774/) (2024)
- [Enhancing Chat Language Models by Scaling High-quality Instructional Conversations](https://mlsanity.com/paper/2305.14233/) (2023)
- [Extending Context Window of Large Language Models via Positional Interpolation](https://mlsanity.com/paper/2306.15595/) (2023)
- [Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunch](https://mlsanity.com/paper/2311.03099/) (2023)
- [To prune, or not to prune: exploring the efficacy of pruning for model compression](https://mlsanity.com/paper/1710.01878/) (2017)
- [Learning multiple visual domains with residual adapters](https://mlsanity.com/paper/1705.08045/) (2017)
- [QuIP: 2-Bit Quantization of Large Language Models With Guarantees](https://mlsanity.com/paper/2307.13304/) (2023)

## Built on by (most cited first, 30 of 39)

- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [Training-Free Activation Sparsity in Large Language Models](https://mlsanity.com/paper/2408.14690/) (2024)
- [Parameter-Efficient Fine-Tuning in Large Models: A Survey of Methodologies](https://mlsanity.com/paper/2410.19878/) (2024)
- [Delta Decompression for MoE-based LLMs Compression](https://mlsanity.com/paper/2502.17298/) (2025)
- [Lottery Ticket Adaptation: Mitigating Destructive Interference in LLMs](https://mlsanity.com/paper/2406.16797/) (2024)
- [Delta-CoMe: Training-Free Delta-Compression with Mixed-Precision for Large Language Models](https://mlsanity.com/paper/2406.08903/) (2024)
- [Personality Vector: Modulating Personality of Large Language Models by Model Merging](https://mlsanity.com/paper/2509.19727/) (2025)
- [FusionBench: A Unified Library and Comprehensive Benchmark for Deep Model Fusion](https://mlsanity.com/paper/2406.03280/) (2024)
- [FREE-Merging: Fourier Transform for Efficient Model Merging](https://mlsanity.com/paper/2411.16815/) (2024)
- [Quantized Delta Weight Is Safety Keeper](https://mlsanity.com/paper/2411.19530/) (2024)
- [DARE the Extreme: Revisiting Delta-Parameter Pruning For Fine-Tuned Models](https://mlsanity.com/paper/2410.09344/) (2024)
- [Knowledge Fusion of Large Language Models Via Modular SkillPacks](https://mlsanity.com/paper/2505.18502/) (2025)
- [Enhancing Perception Capabilities of Multimodal LLMs with Training-Free Fusion](https://mlsanity.com/paper/2412.01289/) (2024)
- [1bit-Merging: Dynamic Quantized Merging for Large Language Models](https://mlsanity.com/paper/2502.10743/) (2025)
- [ImPart: Importance-Aware Delta-Sparsification for Improved Model Compression and Merging in LLMs](https://mlsanity.com/paper/2504.13237/) (2025)
- [Operationalising the Superficial Alignment Hypothesis via Task Complexity](https://mlsanity.com/paper/2602.15829/) (2026)
- [Sketch to Adapt: Fine-Tunable Sketches for Efficient LLM Adaptation](https://mlsanity.com/paper/2410.06364/) (2024)
- [DeltaDQ: Ultra-High Delta Compression for Fine-Tuned LLMs via Group-wise Dropout and Separate Quantization](https://mlsanity.com/paper/2410.08666/) (2024)
- [ME-Switch: A Memory-Efficient Expert Switching Framework for Large Language Models](https://mlsanity.com/paper/2406.09041/) (2024)
- [Novel-WD: Exploring acquisition of Novel World Knowledge in LLMs Using Prefix-Tuning](https://mlsanity.com/paper/2408.17070/) (2024)
- [A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models](https://mlsanity.com/paper/2410.13841/) (2024)
- [Highly Efficient and Effective LLMs with Multi-Boolean Architectures](https://mlsanity.com/paper/2505.22811/) (2025)
- [PoLoRA: A Preconditioned Orthogonalized LoRA Optimizer](https://mlsanity.com/paper/2607.17620/) (2026)
- [Task Vector Bases: A Unified and Scalable Framework for Compressed Task Arithmetic](https://mlsanity.com/paper/2502.01015/) (2025)
- [Seeing Delta Parameters as JPEG Images: Data-Free Delta Compression with Discrete Cosine Transform](https://mlsanity.com/paper/2503.06676/) (2025)
- [Dynamic Base model Shift for Delta Compression](https://mlsanity.com/paper/2505.11344/) (2025)
- [D-QRELO: Training- and Data-Free Delta Compression for Large Language Models via Quantization and Residual Low-Rank Approximation](https://mlsanity.com/paper/2604.16940/) (2026)
- [Dynamic Model Merging Made Slim](https://mlsanity.com/paper/2605.18904/) (2026)
- [Skill Weaving: Efficient LLM Improvement via Modular Skillpacks](https://mlsanity.com/paper/2605.22205/) (2026)
- [In-Cell Learning: Language Models That Update Their Own Weights in Sequence Without Changing the File They Ship](https://mlsanity.com/paper/2608.20873/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are typically trained in two phases: pre-training on large internet-scale datasets, and fine-tuning for downstream tasks. Given the higher computational demand of pre-training, it's intuitive to assume that fine-tuning adds less new information to the model, and is thus more compressible. We explore this assumption by decomposing the weights of fine-tuned models into their pre-trained components and an additional delta. We introduce a simple method, BitDelta, which successfully quantizes this delta down to 1 bit without compromising performance. This interesting finding not only highlights the potential redundancy of information added during fine-tuning, but also has significant implications for the multi-tenant serving and multi-tenant storage of fine-tuned models. By enabling the use of a single high-precision base model accompanied by multiple 1-bit deltas, BitDelta dramatically reduces GPU memory requirements by more than 10x, which can also be translated to enhanced generation latency in multi-tenant settings. We validate BitDelta through experiments across Llama-2 and Mistral model families, and on models up to 70B parameters, showcasing minimal performance degradation over all tested settings.