{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/bitdelta-your-fine-tune-may-only-be-worth-one","title":"BitDelta: Your Fine-Tune May Only Be Worth One Bit","arxiv_id":"2402.10193","date":"2024-02-15","proceeding":null,"authors":["James Liu","Guangxuan Xiao","Kai Li","Jason D. Lee","Song Han","Tri Dao","Tianle Cai"],"abstract":"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.","url_abs":"https://arxiv.org/abs/2402.10193v3","url_pdf":"https://arxiv.org/pdf/2402.10193v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"bitdelta-your-fine-tune-may-only-be-worth-one","repo_url":"https://github.com/FasterDecoding/BitDelta","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":null,"task_name":"GPU"}],"methods":[{"method_slug":"base","method_name":"BASE"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2402.10193","atlas_url":"https://app.syntology.ai/?focus=2402.10193","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2402.10193"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/FasterDecoding/BitDelta","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":5,"ran_draft_wrong":1,"unverified":3},"by_repo_kind":{"official":{"samples":9,"ran":6,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"788def677abe4a95","entry":"find_corr_stddev","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/misc.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/misc.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"788def677abe4a95"}},{"code_sha256_prefix":"c39b5bb17824dd6e","entry":"get_dataloader","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/data.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c39b5bb17824dd6e"}},{"code_sha256_prefix":"3857e2268fd2be5c","entry":"pack","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/binary_gemm_kernel.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/binary_gemm_kernel.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3857e2268fd2be5c"}},{"code_sha256_prefix":"9dcbdcf4fe815b37","entry":"parse_device","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/utils.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/utils.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9dcbdcf4fe815b37"}},{"code_sha256_prefix":"c85ef5aa3bf5041e","entry":"parse_dict","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/utils.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c85ef5aa3bf5041e"}},{"code_sha256_prefix":"7b1afaa65397beea","entry":"unpack","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/binary_gemm_kernel.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/binary_gemm_kernel.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7b1afaa65397beea"}},{"code_sha256_prefix":"ca14e14c001835fe","entry":"dequantize_model","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/misc.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/misc.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ca14e14c001835fe"}},{"code_sha256_prefix":"3ab4b98edc5243cf","entry":"get_dataset","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/data.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3ab4b98edc5243cf"}},{"code_sha256_prefix":"f08bdafbdaf3bee1","entry":"get_model","repo":"FasterDecoding/BitDelta","repo_kind":"official","path":"bitdelta/utils.py","file_url":"https://github.com/FasterDecoding/BitDelta/blob/HEAD/bitdelta/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f08bdafbdaf3bee1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}