{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/method/zero","slug":"zero","name":"ZeRO","full_name":"ZeRO","full_name_withheld":false,"description_markdown":"**Zero Redundancy Optimizer (ZeRO)** is a sharded data parallel method for distributed training. ZeRODP removes the memory state redundancies across data-parallel processes by partitioning the model states instead of replicating them, and it retains the compute/communication efficiency by retaining the computational granularity and communication volume of DP using a dynamic communication schedule during training.","description_state":"present","introduced_year":null,"introduced_by":{"title":"ZeRO: Memory Optimizations Toward Training Trillion Parameter Models","paper":"/paper/zero-memory-optimization-towards-training-a","first_author":"Samyam Rajbhandari","n_authors":4,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/zero-memory-optimization-towards-training-a"},"source":{"url":"https://arxiv.org/abs/1910.02054v3","title":"ZeRO: Memory Optimizations Toward Training Trillion Parameter Models","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Sharded Data Parallel Methods","url":"/methods/category/sharded-data-parallel-methods","pwc_aliases":[]},{"area":"General","area_id":"general","collection":"Data Parallel Methods","url":"/methods/category/data-parallel-methods","pwc_aliases":[]},{"area":"General","area_id":"general","collection":"Distributed Methods","url":"/methods/category/distributed-methods","pwc_aliases":[]}],"n_papers_tagged":9,"archive_num_papers":9,"papers_newest_first":[{"paper":null,"slug":"memory-analysis-on-the-training-course-of","title":"Memory Analysis on the Training Course of DeepSeek Models","date":"2025-02-11","arxiv_id":"2502.07846","n_code_links":0,"syntology":null},{"paper":null,"slug":"accelerating-large-language-model-training","title":"Accelerating Large Language Model Training with Hybrid GPU-based Compression","date":"2024-09-04","arxiv_id":"2409.02423","n_code_links":0,"syntology":null},{"paper":null,"slug":"a-study-of-optimizations-for-fine-tuning","title":"A Study of Optimizations for Fine-tuning Large Language Models","date":"2024-06-04","arxiv_id":"2406.02290","n_code_links":0,"syntology":null},{"paper":null,"slug":"zero-redundancy-distributed-learning-with","title":"Zero redundancy distributed learning with differential privacy","date":"2023-11-20","arxiv_id":"2311.11822","n_code_links":0,"syntology":null},{"paper":null,"slug":"dissecting-the-runtime-performance-of-the","title":"Dissecting the Runtime Performance of the Training, Fine-tuning, and Inference of Large Language Models","date":"2023-11-07","arxiv_id":"2311.03687","n_code_links":0,"syntology":null},{"paper":"/paper/remax-a-simple-effective-and-efficient-method","slug":"remax-a-simple-effective-and-efficient-method","title":"ReMax: A Simple, Effective, and Efficient Reinforcement Learning Method for Aligning Large Language Models","date":"2023-10-16","arxiv_id":"2310.10505","n_code_links":3,"syntology":{"ran":11,"of":13,"n_ran_checked":9,"n_instrument":2,"unverified":2,"pointer_only":13,"phrase":"11 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 1 honoured, 0 violated, 8 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified","official":{"repos":["liziniu/ReMax"],"state":"official (archive's flag): 11 ran","n_ran":11,"n_constructed":0,"n_ran_no_instrument_failure":9,"n_unverified":2,"ran_from_kinds":["official"]}}},{"paper":null,"slug":"rethinking-memory-and-communication-cost-for","title":"Rethinking Memory and Communication Cost for Efficient Large Language Model Training","date":"2023-10-09","arxiv_id":"2310.06003","n_code_links":0,"syntology":null},{"paper":"/paper/zero-extremely-efficient-collective","slug":"zero-extremely-efficient-collective","title":"ZeRO++: Extremely Efficient Collective Communication for Giant Model Training","date":"2023-06-16","arxiv_id":"2306.10209","n_code_links":1,"syntology":null},{"paper":"/paper/zero-memory-optimization-towards-training-a","slug":"zero-memory-optimization-towards-training-a","title":"ZeRO: Memory Optimizations Toward Training Trillion Parameter Models","date":"2019-10-04","arxiv_id":"1910.02054","n_code_links":10,"syntology":{"ran":1,"of":1,"n_ran_checked":1,"n_instrument":0,"unverified":0,"pointer_only":0,"phrase":"1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified","official":{"repos":["NVIDIA/Megatron-LM","microsoft/DeepSpeed"],"state":"community repositories only","n_ran":0,"n_constructed":0,"n_ran_no_instrument_failure":0,"n_unverified":0,"ran_from_kinds":["listed"]}}}],"papers_shown":9,"papers_order":"date (newest first), then slug","paper_counts":{"archive_papers_tagged":9,"with_a_code_link":3,"where_syntology_ran_a_sample":2,"not_listed_spam_title":0,"listed":9,"listed_where_code_ran":2,"where_syntology_ran_a_sample_split":{"with_a_run_with_no_instrument_failure":2,"every_run_a_failure_of_syntologys_instrument":0,"listed_with_a_run_with_no_instrument_failure":2,"listed_every_run_a_failure_of_syntologys_instrument":0,"filter":{"states":["a run with no instrument failure","any run, instrument failures included"],"default":"a run with no instrument failure","note":"on the 'only where code ran' pages the default hides, in the browser, the rows where every run was a failure of Syntology's instrument; the second state shows them again. Rows are hidden, never re-ordered; these twins list every row"}},"definition":"distinct papers the archive tags; 'where Syntology ran a sample' counts papers with at least one harvested sample that ran, which is not a correctness claim"},"papers_where_code_ran":2,"papers_where_code_ran_first_page":"/method/zero/papers/ran/1","papers_where_code_ran_caption":"We ran code from the paper's repository; we did not isolate this method inside it.","pages":[{"order":"ran","page":1,"of":1,"url":"/method/zero/papers/ran/1","json":"/method/zero/papers/ran/1.json","rows":2}],"tasks":[{"task":null,"name":"GPU","papers":7},{"task":"/task/language-modelling","name":"Language Modelling","papers":3},{"task":"/task/language-modeling","name":"Language Modeling","papers":2},{"task":"/task/large-language-model","name":"Large Language Model","papers":2},{"task":"/task/quantization","name":"Quantization","papers":2},{"task":"/task/cross-lingual-document-classification","name":"Cross-Lingual Document Classification","papers":1},{"task":"/task/general-reinforcement-learning","name":"General Reinforcement Learning","papers":1},{"task":"/task/image-generation","name":"Image Generation","papers":1},{"task":"/task/mixture-of-experts","name":"Mixture-of-Experts","papers":1},{"task":"/task/privacy-preserving","name":"Privacy Preserving","papers":1},{"task":"/task/reinforcement-learning-2","name":"reinforcement-learning","papers":1}],"tasks_shown":11,"n_tasks":11,"usage_by_year":[{"year":"2019","papers":1},{"year":"2023","papers":5},{"year":"2024","papers":2},{"year":"2025","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/zero"},"syntology_read_at":"2026-10-05T00:36:11+00:00","syntology":{"source":"Syntology's have check for this name (the free coverage lookup), read from Syntology's graph","have_url":"https://syntology.ai/api/graph/have?x=ZeRO","read_at":"2026-10-05T01:04:04+00:00","query":"ZeRO","kind":"method","have":"declared_code","tier":"repo_known","verified_implementation":false,"level":null,"level_meaning":"the served verification level, per answer; never summed across methods","resolved_to":{"method":"ZERO","origin_arxiv_id":"2405.18330"},"resolution":"case_insensitive","resolved_name_differs":false,"absence_asserted":false,"absence_unchecked":null,"no_code_is_about_this_method":false,"subject":"different_paper","archive_paper_arxiv_id":"1910.02054","repositories":{"listed":1,"in_graph":1,"list_truncated":false,"strongest_link":"official"},"retrieve_with":null,"name_competition":{"verdict":"outranked","may_be_wrong_sense":true,"bearers_found":3,"served":{"arxiv_id":"2405.18330","citations":41},"leading_other_bearer":{"arxiv_id":"1910.02054","title":"ZeRO: Memory Optimizations Toward Training Trillion Parameter Models","citations":2002,"named_by":["title_prefix"],"bearer_name":"ZeRO: Memory Optimizations Toward Training Trillion Parameter Models","ask_with":"have('1910.02054')"},"n_other_bearers_listed":2,"note":"'ZeRO' is ALSO the name of the paper 'ZeRO: Memory Optimizations Toward Training Trillion Parameter Models' (arXiv:1910.02054, 2,002 citations), which is cited FAR more than the paper this answer is about (41). This answer may be about the wrong ZeRO. If you meant the better-known one, ask again with that entry of `also_named`."},"name_competition_text":"The same name also belongs to the paper “ZeRO: Memory Optimizations Toward Training Trillion Parameter Models” (arXiv:1910.02054, 2,002 citations), which the lookup finds cited far more than the paper this answer is about (41 citations), so this answer may be about a different sense of the name.","near_matches":null,"code_nearby":[{"name":"Zero-Shot Tokenizer Transfer (ZeTT)","relation":"shares_name_tokens","relation_means":"shares name tokens with this name; a different method","level":2,"retrieve_with":"get_reference_implementation(name='Zero-Shot Tokenizer Transfer (ZeTT)')","origin_arxiv_id":"2405.07883"},{"name":"Simple Zero-shot Prompt Weighting Technique","relation":"shares_name_tokens","relation_means":"shares name tokens with this name; a different method","level":3,"retrieve_with":"get_reference_implementation(name='Simple Zero-shot Prompt Weighting Technique')","origin_arxiv_id":"2302.06235"}],"code_nearby_caption":"Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.","origin_paper":{"arxiv_id":"2405.18330","page":"/paper/frustratingly-easy-test-time-adaptation-of","syntology_url":"https://syntology.ai/paper/2405.18330","in_graph_layer":true,"samples":8,"ran":4,"n_ran_checked":1,"n_instrument":3,"unverified":4,"phrase":"4 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 3 where Syntology's instrument failed) · 4 unverified","instrument_means":"failures of Syntology's instrument, not of the code"},"line":"about another paper, arXiv:2405.18330: repository known, no verified implementation (declared_code; repositories: 1 listed of 1 in the graph, strongest link: official). The lookup matched 'ZERO' ignoring case. About another paper: this answer is about arXiv:2405.18330, while the archive credits arXiv:1910.02054.","level_legend":{"state":"quoted","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","card_sha256":"479785a84658dd682a11b10237b3620d651d7f87799f5dc7add970aaa83f6d39","note":"Syntology's served verification levels, quoted verbatim from its MCP server card; a row is split where the card separates clauses with a double hyphen, and the separator is not reproduced","levels":[{"level":"V0","quoted":["imports"]},{"level":"V1","quoted":["executes against a synthesized fixture and returns a value of the recorded type"]},{"level":"V2","quoted":["two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs","neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)"]},{"level":"V3","quoted":["paper-derived property tests pass on both of those same two generations."]}],"no_rung":"No rung on this ladder compares the served code against the authors' implementation, so treat every level as evidence about runnability and internal agreement, not about matching the paper's results."}},"syntology_have_state":"loaded","record_sha256":"1103ed979bcb86b3a2f5425e822b3053e5b5ec7fbeb5a6364b8204b904fa4274","record_changed_at":"2026-10-05","record_changed_at_basis":"changed"}