{"id":"https://openalex.org/W7225790884","doi":"https://doi.org/10.5281/zenodo.24136832","title":"Zorlo: Running Large Mixture-of-Experts Models on Low-RAM PCs by Streaming Experts from the SSD","display_name":"Zorlo: Running Large Mixture-of-Experts Models on Low-RAM PCs by Streaming Experts from the SSD","publication_year":2026,"publication_date":"2026-11-11","ids":{"openalex":"https://openalex.org/W7225790884","doi":"https://doi.org/10.5281/zenodo.24136832"},"language":"en","primary_location":{"id":"doi:10.5281/zenodo.24136832","doi":"https://doi.org/10.5281/zenodo.24136832","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24136832","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article-journal"},"volume":null,"type":"article","is_traditional":true,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.5281/zenodo.24136832","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5161289204","display_name":"Haider Md. Faiyaz Hussain Shahi","orcid":"https://orcid.org/0009-0004-0940-9096","observed_orcids":["https://orcid.org/0009-0004-0940-9096"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Shahi, Haider Md. Faiyaz Hussain","raw_affiliation_strings":["zorlo"],"raw_orcid":"https://orcid.org/0009-0004-0940-9096","affiliations":[{"raw_affiliation_string":"zorlo","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5161289204"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.0,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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":0.4350999891757965,"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":0.4350999891757965,"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/T10028","display_name":"Topic Modeling","score":0.2687999904155731,"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.12080000340938568,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7942000031471252},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6316999793052673},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.6175000071525574},{"id":"https://openalex.org/C2780008327","wikidata":"https://www.wikidata.org/wiki/Q3962","display_name":"Laptop","level":2,"score":0.6026999950408936},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.593500018119812},{"id":"https://openalex.org/C2778858076","wikidata":"https://www.wikidata.org/wiki/Q5249539","display_name":"Decodes","level":3,"score":0.5889999866485596},{"id":"https://openalex.org/C554936623","wikidata":"https://www.wikidata.org/wiki/Q199657","display_name":"Reading (process)","level":2,"score":0.5307000279426575},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.3813000023365021},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.35109999775886536},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3433000147342682},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2948000133037567},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.2913999855518341},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.26739999651908875},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.26159998774528503}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.5281/zenodo.24136832","doi":"https://doi.org/10.5281/zenodo.24136832","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24136832","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article-journal"},{"id":"doi:10.5281/zenodo.24136833","doi":"https://doi.org/10.5281/zenodo.24136833","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24136833","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article-journal"}],"best_oa_location":{"id":"doi:10.5281/zenodo.24136832","doi":"https://doi.org/10.5281/zenodo.24136832","is_oa":true,"landing_page_url":"https://doi.org/10.5281/zenodo.24136832","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article-journal"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Mixture-of-Experts":[0],"(MoE)":[1],"language":[2],"models":[3,51],"are":[4,79],"large":[5],"on":[6,52,75,81,140,187],"disk":[7],"but":[8],"use":[9],"only":[10,131],"a":[11,84,97,115,128,169,200],"small":[12],"part":[13],"of":[14,103,130],"their":[15],"weights":[16,60],"for":[17,42,70],"any":[18],"one":[19],"token.":[20,33],"Qwen3-30B-A3B":[21,107],"stores":[22],"about":[23,29,109],"30":[24],"billion":[25,31],"parameters":[26],"and":[27,39,78,122,151,208],"activates":[28],"3":[30],"per":[32,112,138,166,185],"Zorlo":[34,105],"is":[35,155],"an":[36],"inference":[37],"engine":[38,175],"desktop":[40],"app":[41],"Windows":[43],"that":[44,61,86,188],"uses":[45],"this":[46],"sparsity":[47],"to":[48,89,117,125,158,182],"run":[49],"such":[50],"ordinary":[53],"low-RAM":[54],"PCs":[55],"with":[56,100,114,127,168],"no":[57],"GPU.":[58],"The":[59,72,173],"every":[62],"token":[63],"needs":[64],"stay":[65,74],"in":[66],"RAM":[67],"(0.82":[68],"GB":[69,102,119],"Qwen3-30B-A3B).":[71],"experts":[73],"the":[76,90,92,141,145,177,197,203,209],"SSD":[77],"read":[80],"demand":[82],"into":[83],"cache":[85,129],"sizes":[87],"itself":[88],"memory":[91,149],"PC":[93],"has":[94],"free.":[95],"On":[96],"4-core":[98],"laptop":[99],"8":[101],"RAM,":[104],"decodes":[106],"at":[108,123,180,199],"3.4":[110],"tokens":[111,137,165,184],"second":[113,139,167,186],"1.2":[116],"1.5":[118,181],"expert":[120],"cache,":[121],"2.3":[124],"3.0":[126],"256":[132],"MB.":[133],"llama.cpp":[134,191],"reaches":[135],"1.43":[136],"same":[142,146,174],"machine,":[143],"reading":[144],"file":[147],"through":[148],"mapping,":[150],"Zorlo's":[152],"default":[153],"output":[154],"token-for-token":[156],"identical":[157],"llama.cpp's.":[159],"Optional":[160],"speed":[161],"modes":[162],"reach":[163],"5.5":[164],"measured":[170],"quality":[171],"cost.":[172],"runs":[176],"117-billion-parameter":[178],"GPT-OSS-120B":[179],"1.7":[183],"laptop,":[189],"where":[190],"manages":[192],"0.37.":[193],"This":[194],"report":[195],"describes":[196],"design":[198],"high":[201],"level,":[202],"measurements":[204],"behind":[205],"these":[206],"numbers,":[207],"current":[210],"limits.":[211]},"counts_by_year":[],"updated_date":"2026-11-12T12:28:57.059296","created_date":"2026-11-12T00:00:00"}
