{"id":"https://openalex.org/W7224084965","doi":"https://doi.org/10.48550/arxiv.2610.36564","title":"DeepInfer-Mac: Running a 750B-Class MoE on a 24 GB Laptop","display_name":"DeepInfer-Mac: Running a 750B-Class MoE on a 24 GB Laptop","publication_year":2026,"publication_date":"2026-10-27","ids":{"openalex":"https://openalex.org/W7224084965","doi":"https://doi.org/10.48550/arxiv.2610.36564"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2610.36564","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.36564","pdf_url":"https://arxiv.org/pdf/2610.36564","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/2610.36564","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5160014493","display_name":"Haorui Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Li, Haorui","raw_affiliation_strings":["Southeast University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":["https://openalex.org/I76569877"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5160014493"],"corresponding_institution_ids":["https://openalex.org/I76569877"],"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8628000020980835,"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"}},"topics":[{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.8628000020980835,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.027699999511241913,"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":0.027400000020861626,"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/deepseek-v4","display_name":"DeepSeek-V4","score":0.6039999723434448},{"id":"https://openalex.org/keywords/large-language-model-inference","display_name":"large language model inference","score":0.609000027179718},{"id":"https://openalex.org/keywords/low-bit-quantization","display_name":"low-bit quantization","score":0.6600000262260437},{"id":"https://openalex.org/keywords/memory-efficient-inference","display_name":"memory-efficient inference","score":0.43700000643730164},{"id":"https://openalex.org/keywords/unified-memory","display_name":"unified memory","score":0.4339999854564667},{"id":"https://openalex.org/keywords/code-generation","display_name":"code generation","score":0.45100000500679016}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7050999999046326},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.6414999961853027},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5716000199317932},{"id":"https://openalex.org/C2780008327","wikidata":"https://www.wikidata.org/wiki/Q3962","display_name":"Laptop","level":2,"score":0.5561000108718872},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.48489999771118164},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.42809998989105225},{"id":"https://openalex.org/C135598885","wikidata":"https://www.wikidata.org/wiki/Q1366302","display_name":"Row","level":2,"score":0.3873000144958496},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.38440001010894775},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3693999946117401},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.36559998989105225},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3440999984741211},{"id":"https://openalex.org/C132943942","wikidata":"https://www.wikidata.org/wiki/Q2562511","display_name":"Footprint","level":2,"score":0.3327000141143799},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.32910001277923584},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3240000009536743},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.32120001316070557},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.30399999022483826},{"id":"https://openalex.org/C194739806","wikidata":"https://www.wikidata.org/wiki/Q66221","display_name":"Computer data storage","level":2,"score":0.29899999499320984},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2953999936580658},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.2872999906539917},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.2816999852657318},{"id":"https://openalex.org/C176649486","wikidata":"https://www.wikidata.org/wiki/Q2308807","display_name":"Memory management","level":3,"score":0.2583000063896179},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.25369998812675476},{"id":"https://openalex.org/C162372511","wikidata":"https://www.wikidata.org/wiki/Q218341","display_name":"Checksum","level":2,"score":0.2515999972820282}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2610.36564","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.36564","pdf_url":"https://arxiv.org/pdf/2610.36564","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.2610.36564","doi":"https://doi.org/10.48550/arxiv.2610.36564","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2610.36564","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2610.36564","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2610.36564","pdf_url":"https://arxiv.org/pdf/2610.36564","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/W7224084965.pdf","grobid_xml":"https://content.openalex.org/works/W7224084965.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7167834298","https://openalex.org/W7171240696","https://openalex.org/W7168247335","https://openalex.org/W7216029924","https://openalex.org/W7171495554","https://openalex.org/W7208818919","https://openalex.org/W7162643580","https://openalex.org/W7164189829","https://openalex.org/W7171299342","https://openalex.org/W7214309931"],"abstract_inverted_index":{"DeepInfer-Mac":[0],"runs":[1],"the":[2,24,46,79,87,109,128,139,142,149,170],"750B-class":[3],"DeepSeek-V4.1-Flash":[4],"model":[5,47,156],"on":[6,23],"a":[7,57,100,116,155],"MacBook":[8],"Air":[9],"with":[10,78,99],"24":[11],"GB":[12],"of":[13,30,59,81,105,130,151],"unified":[14],"memory.":[15],"A":[16],"mixed":[17],"low-bit":[18],"conversion":[19],"occupies":[20],"338":[21],"GiB":[22,29,107],"internal":[25],"SSD.":[26],"About":[27],"7.6":[28],"weights":[31,65],"remain":[32,135],"resident;":[33],"routed":[34],"experts":[35],"and":[36,74,92,144,147,166],"Engram":[37],"rows":[38],"are":[39],"loaded":[40],"when":[41],"needed.":[42],"This":[43],"separation":[44],"makes":[45],"executable,":[48],"but":[49],"it":[50],"leaves":[51],"each":[52],"generated":[53],"token":[54],"waiting":[55],"for":[56],"succession":[58],"storage":[60,165],"reads.":[61,132],"We":[62,137],"organize":[63],"expert":[64],"so":[66],"that":[67],"reads":[68],"fill":[69],"reusable":[70],"GPU":[71],"buffers":[72],"directly,":[73],"couple":[75],"cache":[76],"allocation":[77],"lifetime":[80],"those":[82],"buffers.":[83],"In":[84],"end-to-end":[85],"runs,":[86],"system":[88],"generates":[89],"English,":[90],"Chinese":[91],"code":[93],"at":[94,127],"0.21-0.72":[95],"tokens":[96],"per":[97],"second,":[98],"largest":[101],"sampled":[102],"process":[103],"footprint":[104],"16.73":[106],"across":[108],"tested":[110],"requests.":[111],"Matched":[112],"decode":[113],"comparisons":[114],"show":[115],"21.0%":[117],"throughput":[118],"gain":[119],"from":[120],"next-layer":[121],"prefetching":[122],"over":[123],"resident-expert":[124],"overlap":[125],"alone,":[126],"cost":[129],"additional":[131],"Longer":[133],"prompts":[134],"costly.":[136],"describe":[138],"design,":[140],"provide":[141],"implementation":[143],"measurement":[145],"scripts,":[146],"identify":[148],"limits":[150],"this":[152],"operating":[153],"point:":[154],"far":[157],"larger":[158],"than":[159],"memory":[160],"can":[161],"run":[162],"locally,":[163],"while":[164],"prefill":[167],"still":[168],"govern":[169],"user's":[171],"wait.":[172]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-10-29T00:00:00"}