{"id":"https://openalex.org/W4416544157","doi":"https://doi.org/10.48550/arxiv.2504.12285","title":"BitNet b1.58 2B4T Technical Report","display_name":"BitNet b1.58 2B4T Technical Report","publication_year":2025,"publication_date":"2025-04-16","ids":{"openalex":"https://openalex.org/W4416544157","doi":"https://doi.org/10.48550/arxiv.2504.12285"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2504.12285","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.12285","pdf_url":"https://arxiv.org/pdf/2504.12285","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2504.12285","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5113130010","display_name":"Shuming Ma","orcid":"https://orcid.org/0000-0003-1091-1206"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Ma, Shuming","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100422655","display_name":"Hongyu Wang","orcid":"https://orcid.org/0000-0003-0031-837X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Hongyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061624006","display_name":"Shaohan Huang","orcid":"https://orcid.org/0000-0003-4324-6337"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Shaohan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120513432","display_name":"Xingxing Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xingxing","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050578796","display_name":"Ying Jun Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Ying","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055341182","display_name":"Ting Song","orcid":"https://orcid.org/0000-0003-3292-7698"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Ting","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045910576","display_name":"Yan Xia","orcid":"https://orcid.org/0000-0003-3121-4784"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xia, Yan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5014662947","display_name":"Furu Wei","orcid":"https://orcid.org/0000-0002-7810-5852"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Furu","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5113130010"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.21809999644756317,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.21809999644756317,"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":0.12610000371932983,"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"}},{"id":"https://openalex.org/T14347","display_name":"Big Data and Digital Economy","score":0.0982000008225441,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/decoding-methods","display_name":"Decoding methods","score":0.5435000061988831},{"id":"https://openalex.org/keywords/implementation","display_name":"Implementation","score":0.5024999976158142},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.49880000948905945},{"id":"https://openalex.org/keywords/coding","display_name":"Coding (social sciences)","score":0.4555000066757202},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.420199990272522},{"id":"https://openalex.org/keywords/face","display_name":"Face (sociological concept)","score":0.3797000050544739}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6625000238418579},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.5435000061988831},{"id":"https://openalex.org/C26713055","wikidata":"https://www.wikidata.org/wiki/Q245962","display_name":"Implementation","level":2,"score":0.5024999976158142},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.49880000948905945},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.4555000066757202},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.420199990272522},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.41940000653266907},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3970000147819519},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38260000944137573},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.3797000050544739},{"id":"https://openalex.org/C76969082","wikidata":"https://www.wikidata.org/wiki/Q486902","display_name":"Mathematical model","level":2,"score":0.32440000772476196},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.31610000133514404},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.3043000102043152},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.2924000024795532},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.2768000066280365},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.27649998664855957},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.27090001106262207},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26159998774528503},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.26109999418258667},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2531000077724457}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2504.12285","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.12285","pdf_url":"https://arxiv.org/pdf/2504.12285","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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.2504.12285","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.12285","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2504.12285","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.12285","pdf_url":"https://arxiv.org/pdf/2504.12285","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416544157.pdf","grobid_xml":"https://content.openalex.org/works/W4416544157.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"introduce":[1],"BitNet":[2,49],"b1.58":[3,50],"2B4T,":[4],"the":[5,15,27,87],"first":[6],"open-source,":[7],"native":[8],"1-bit":[9],"Large":[10],"Language":[11],"Model":[12],"(LLM)":[13],"at":[14],"2-billion":[16],"parameter":[17],"scale.":[18],"Trained":[19],"on":[20,54],"a":[21],"corpus":[22],"of":[23,61],"4":[24],"trillion":[25],"tokens,":[26],"model":[28,88],"has":[29],"been":[30],"rigorously":[31],"evaluated":[32],"across":[33],"benchmarks":[34],"covering":[35],"language":[36],"understanding,":[37],"mathematical":[38],"reasoning,":[39],"coding":[40],"proficiency,":[41],"and":[42,78,85,103],"conversational":[43],"ability.":[44],"Our":[45],"results":[46],"demonstrate":[47],"that":[48],"2B4T":[51],"achieves":[52],"performance":[53],"par":[55],"with":[56,96],"leading":[57],"open-weight,":[58],"full-precision":[59],"LLMs":[60],"similar":[62],"size,":[63],"while":[64],"offering":[65],"significant":[66],"advantages":[67],"in":[68],"computational":[69],"efficiency,":[70],"including":[71],"substantially":[72],"reduced":[73],"memory":[74],"footprint,":[75],"energy":[76],"consumption,":[77],"decoding":[79],"latency.":[80],"To":[81],"facilitate":[82],"further":[83],"research":[84],"adoption,":[86],"weights":[89],"are":[90],"released":[91],"via":[92],"Hugging":[93],"Face":[94],"along":[95],"open-source":[97],"inference":[98],"implementations":[99],"for":[100],"both":[101],"GPU":[102],"CPU":[104],"architectures.":[105]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-10-10T00:00:00"}