{"id":"https://openalex.org/W4404986232","doi":"https://doi.org/10.48550/arxiv.2411.15242","title":"The Zamba2 Suite: Technical Report","display_name":"The Zamba2 Suite: Technical Report","publication_year":2024,"publication_date":"2024-11-22","ids":{"openalex":"https://openalex.org/W4404986232","doi":"https://doi.org/10.48550/arxiv.2411.15242"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2411.15242","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2411.15242","pdf_url":"https://arxiv.org/pdf/2411.15242","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2411.15242","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5027899526","display_name":"Paolo Glorioso","orcid":"https://orcid.org/0000-0002-5211-2982","observed_orcids":["https://orcid.org/0000-0002-5211-2982"]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Glorioso, Paolo","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015153122","display_name":"Quentin G. Anthony","orcid":"https://orcid.org/0000-0002-6823-9080","observed_orcids":["https://orcid.org/0000-0002-6823-9080"]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anthony, Quentin","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093876166","display_name":"Yury Tokpanov","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tokpanov, Yury","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114987484","display_name":"Anna Golubeva","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Golubeva, Anna","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5035178674","display_name":"Vasudev Shyam","orcid":"https://orcid.org/0000-0003-2037-941X","observed_orcids":["https://orcid.org/0000-0003-2037-941X"]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shyam, Vasudev","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5098921489","display_name":"James Whittington","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Whittington, James","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5049628803","display_name":"Jonathan Pilault","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Pilault, Jonathan","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5048239223","display_name":"Beren Millidge","orcid":"https://orcid.org/0000-0003-1872-5635","observed_orcids":["https://orcid.org/0000-0003-1872-5635"]},"institutions":[{"id":"https://openalex.org/I4405265938","display_name":"Zyphra (United States)","ror":"https://ror.org/01wzpra32","country_code":"US","type":"company","lineage":["https://openalex.org/I4405265938"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Millidge, Beren","raw_affiliation_strings":["Zyphra Palo Alto , CA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zyphra Palo Alto , CA","institution_ids":["https://openalex.org/I4405265938"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4405265938"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":8,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.8833000063896179,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.8833000063896179,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.060100000351667404,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.019600000232458115,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/mamba-2","display_name":"Mamba-2","score":0.6700000166893005},{"id":"https://openalex.org/keywords/state-space-models","display_name":"state-space model","score":0.8320000171661377},{"id":"https://openalex.org/keywords/open-weights","display_name":"open weights","score":0.4309999942779541},{"id":"https://openalex.org/keywords/inference-latency","display_name":"inference latency","score":0.5540000200271606},{"id":"https://openalex.org/keywords/memory-efficiency","display_name":"memory efficiency","score":0.9179999828338623},{"id":"https://openalex.org/keywords/instruction-tuning","display_name":"instruction tuning","score":0.6779999732971191},{"id":"https://openalex.org/keywords/long-sequence-modeling","display_name":"long sequence modeling","score":0.3319999873638153}],"concepts":[{"id":"https://openalex.org/C79581498","wikidata":"https://www.wikidata.org/wiki/Q1367530","display_name":"Suite","level":2,"score":0.8986256122589111},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4372675120830536},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.1823568344116211},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0903271734714508}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2411.15242","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2411.15242","pdf_url":"https://arxiv.org/pdf/2411.15242","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2411.15242","doi":"https://doi.org/10.48550/arxiv.2411.15242","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2411.15242","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2411.15242","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2411.15242","pdf_url":"https://arxiv.org/pdf/2411.15242","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4404986232.pdf","grobid_xml":"https://content.openalex.org/works/W4404986232.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4399115929","https://openalex.org/W4416053115","https://openalex.org/W4405664040","https://openalex.org/W7154499146","https://openalex.org/W7222585852","https://openalex.org/W4403346143","https://openalex.org/W4399657245","https://openalex.org/W4415797412","https://openalex.org/W4416050145","https://openalex.org/W4393399080"],"abstract_inverted_index":{"In":[0],"this":[1,123],"technical":[2],"report,":[3],"we":[4,106],"present":[5],"the":[6,25,29,80,102,112],"Zamba2":[7,48,81,113],"series":[8,49,82,114],"--":[9],"a":[10],"suite":[11],"of":[12,24,33,79,96,115],"1.2B,":[13],"2.7B,":[14],"and":[15,44,61,64,119],"7.4B":[16],"parameter":[17],"hybrid":[18],"Mamba2-transformer":[19],"models":[20,32,78,95,118],"that":[21,88],"achieve":[22],"state":[23],"art":[26],"performance":[27],"against":[28,92],"leading":[30],"open-weights":[31],"their":[34,97],"class,":[35],"while":[36],"achieving":[37],"substantial":[38],"gains":[39],"in":[40,122],"inference":[41],"latency,":[42],"throughput,":[43],"memory":[45],"efficiency.":[46],"The":[47,117],"builds":[50],"upon":[51],"our":[52],"initial":[53],"work":[54,124],"with":[55],"Zamba1-7B,":[56],"optimizing":[57],"its":[58],"architecture,":[59],"training":[60,65],"annealing":[62],"datasets,":[63],"for":[66,76],"up":[67],"to":[68,110],"three":[69],"trillion":[70],"tokens.":[71],"We":[72,99],"provide":[73],"open-source":[74,101],"weights":[75],"all":[77],"as":[83,85],"well":[84],"instruction-tuned":[86],"variants":[87],"are":[89,125],"strongly":[90],"competitive":[91],"comparable":[93],"instruct-tuned":[94],"class.":[98],"additionally":[100],"pretraining":[103],"dataset,":[104],"which":[105],"call":[107],"Zyda-2,":[108],"used":[109,121],"train":[111],"models.":[116],"datasets":[120],"openly":[126],"available":[127],"at":[128],"https://huggingface.co/Zyphra":[129]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":4}],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2024-12-04T00:00:00"}
