{"id":"https://openalex.org/W7139046211","doi":"https://doi.org/10.1016/j.future.2026.108483","title":"Energy-efficient large language models","display_name":"Energy-efficient large language models","publication_year":2026,"publication_date":"2026-03-19","ids":{"openalex":"https://openalex.org/W7139046211","doi":"https://doi.org/10.1016/j.future.2026.108483"},"language":"en","primary_location":{"id":"doi:10.1016/j.future.2026.108483","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.future.2026.108483","pdf_url":null,"source":{"id":"https://openalex.org/S186357190","display_name":"Future Generation Computer Systems","issn_l":"0167-739X","issn":["0167-739X","1872-7115"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Future Generation Computer Systems","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1016/j.future.2026.108483","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5024407020","display_name":"Javier Jare\u00f1o","orcid":"https://orcid.org/0000-0002-1735-0520"},"institutions":[{"id":"https://openalex.org/I39147953","display_name":"Universidad de C\u00e1diz","ror":"https://ror.org/04mxxkb11","country_code":"ES","type":"education","lineage":["https://openalex.org/I39147953"]}],"countries":["ES"],"is_corresponding":true,"raw_author_name":"Javier Jare\u00f1o","raw_affiliation_strings":["Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain"],"raw_orcid":"https://orcid.org/0000-0002-1735-0520","affiliations":[{"raw_affiliation_string":"Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain","institution_ids":["https://openalex.org/I39147953"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012267701","display_name":"Jos\u00e9 Miguel Arag\u00f3n-Jurado","orcid":"https://orcid.org/0000-0002-9835-8793"},"institutions":[{"id":"https://openalex.org/I39147953","display_name":"Universidad de C\u00e1diz","ror":"https://ror.org/04mxxkb11","country_code":"ES","type":"education","lineage":["https://openalex.org/I39147953"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Jos\u00e9 Miguel Arag\u00f3n-Jurado","raw_affiliation_strings":["Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain"],"raw_orcid":"https://orcid.org/0000-0002-9835-8793","affiliations":[{"raw_affiliation_string":"Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain","institution_ids":["https://openalex.org/I39147953"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064449004","display_name":"Juan Carlos de la Torre","orcid":"https://orcid.org/0000-0002-3662-6422"},"institutions":[{"id":"https://openalex.org/I39147953","display_name":"Universidad de C\u00e1diz","ror":"https://ror.org/04mxxkb11","country_code":"ES","type":"education","lineage":["https://openalex.org/I39147953"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Juan Carlos De La Torre","raw_affiliation_strings":["Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain"],"raw_orcid":"https://orcid.org/0000-0002-3662-6422","affiliations":[{"raw_affiliation_string":"Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain","institution_ids":["https://openalex.org/I39147953"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000488528","display_name":"Patricia Ruiz","orcid":"https://orcid.org/0000-0001-5891-0646"},"institutions":[{"id":"https://openalex.org/I39147953","display_name":"Universidad de C\u00e1diz","ror":"https://ror.org/04mxxkb11","country_code":"ES","type":"education","lineage":["https://openalex.org/I39147953"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Patricia Ruiz","raw_affiliation_strings":["Department of Mechanical Engineering and Industrial Design School of Engineering, University of Cadiz, Spain"],"raw_orcid":"https://orcid.org/0000-0001-5891-0646","affiliations":[{"raw_affiliation_string":"Department of Mechanical Engineering and Industrial Design School of Engineering, University of Cadiz, Spain","institution_ids":["https://openalex.org/I39147953"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061059522","display_name":"Bernab\u00e8 Dorronsoro","orcid":"https://orcid.org/0000-0003-0481-790X"},"institutions":[{"id":"https://openalex.org/I39147953","display_name":"Universidad de C\u00e1diz","ror":"https://ror.org/04mxxkb11","country_code":"ES","type":"education","lineage":["https://openalex.org/I39147953"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Bernab\u00e9 Dorronsoro","raw_affiliation_strings":["Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain"],"raw_orcid":"https://orcid.org/0000-0003-0481-790X","affiliations":[{"raw_affiliation_string":"Department of Computer Science Engineering, School of Engineering, University of Cadiz, Spain","institution_ids":["https://openalex.org/I39147953"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":["https://openalex.org/A5024407020"],"corresponding_institution_ids":["https://openalex.org/I39147953"],"apc_list":{"value":3340,"currency":"USD","value_usd":3340},"apc_paid":{"value":3340,"currency":"USD","value_usd":3340},"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.33922241,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"182","issue":null,"first_page":"108483","last_page":"108483"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T14347","display_name":"Big Data and Digital Economy","score":0.1412999927997589,"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"}},"topics":[{"id":"https://openalex.org/T14347","display_name":"Big Data and Digital Economy","score":0.1412999927997589,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.1185000017285347,"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":0.0658000037074089,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/energy-consumption","display_name":"Energy consumption","score":0.6044999957084656},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5820000171661377},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5794000029563904},{"id":"https://openalex.org/keywords/genetic-algorithm","display_name":"Genetic algorithm","score":0.5266000032424927},{"id":"https://openalex.org/keywords/crossover","display_name":"Crossover","score":0.43950000405311584},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.4352000057697296},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.41920000314712524},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.4171000123023987},{"id":"https://openalex.org/keywords/energy","display_name":"Energy (signal processing)","score":0.39989998936653137}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8791999816894531},{"id":"https://openalex.org/C2780165032","wikidata":"https://www.wikidata.org/wiki/Q16869822","display_name":"Energy consumption","level":2,"score":0.6044999957084656},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5820000171661377},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5794000029563904},{"id":"https://openalex.org/C8880873","wikidata":"https://www.wikidata.org/wiki/Q187787","display_name":"Genetic algorithm","level":2,"score":0.5266000032424927},{"id":"https://openalex.org/C122507166","wikidata":"https://www.wikidata.org/wiki/Q628906","display_name":"Crossover","level":2,"score":0.43950000405311584},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.4352000057697296},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.41920000314712524},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.4171000123023987},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.39989998936653137},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3946000039577484},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37790000438690186},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.3709000051021576},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.36649999022483826},{"id":"https://openalex.org/C14961307","wikidata":"https://www.wikidata.org/wiki/Q5377176","display_name":"Energy minimization","level":2,"score":0.36480000615119934},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3580999970436096},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3555999994277954},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.35350000858306885},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34599998593330383},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3398999869823456},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.3375999927520752},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.313400000333786},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3098999857902527},{"id":"https://openalex.org/C501734568","wikidata":"https://www.wikidata.org/wiki/Q42918","display_name":"Mutation","level":3,"score":0.29429998993873596},{"id":"https://openalex.org/C137877099","wikidata":"https://www.wikidata.org/wiki/Q1332977","display_name":"Subsequence","level":3,"score":0.28949999809265137},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.2793999910354614},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2736999988555908},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C51823790","wikidata":"https://www.wikidata.org/wiki/Q504353","display_name":"Greedy algorithm","level":2,"score":0.26739999651908875},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.26420000195503235},{"id":"https://openalex.org/C111696304","wikidata":"https://www.wikidata.org/wiki/Q2303697","display_name":"Sorting","level":2,"score":0.25360000133514404}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1016/j.future.2026.108483","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.future.2026.108483","pdf_url":null,"source":{"id":"https://openalex.org/S186357190","display_name":"Future Generation Computer Systems","issn_l":"0167-739X","issn":["0167-739X","1872-7115"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Future Generation Computer Systems","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1016/j.future.2026.108483","is_oa":true,"landing_page_url":"https://doi.org/10.1016/j.future.2026.108483","pdf_url":null,"source":{"id":"https://openalex.org/S186357190","display_name":"Future Generation Computer Systems","issn_l":"0167-739X","issn":["0167-739X","1872-7115"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320990","host_organization_name":"Elsevier BV","host_organization_lineage":["https://openalex.org/P4310320990"],"host_organization_lineage_names":["Elsevier BV"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Future Generation Computer Systems","raw_type":"journal-article"},"sustainable_development_goals":[{"score":0.8177732229232788,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[{"id":"https://openalex.org/G1108718739","display_name":null,"funder_award_id":"FPU24/04054","funder_id":"https://openalex.org/F4320315062","funder_display_name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades"},{"id":"https://openalex.org/G3443813095","display_name":null,"funder_award_id":"PIDI_2024_01307","funder_id":"https://openalex.org/F3619780700","funder_display_name":"Consejer\u00eda de Universidad, Investigaci\u00f3n e Innovaci\u00f3n, Junta de Andaluc\u00eda"},{"id":"https://openalex.org/G3970630980","display_name":null,"funder_award_id":"PID2022-137858OB-I00","funder_id":"https://openalex.org/F4320315062","funder_display_name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades"},{"id":"https://openalex.org/G5898280210","display_name":null,"funder_award_id":"DGP PRED 2024 00654","funder_id":"https://openalex.org/F4320319605","funder_display_name":"Consejer\u00eda de Transformaci\u00f3n Econ\u00f3mica, Industria, Conocimiento y Universidades"},{"id":"https://openalex.org/G6169016097","display_name":null,"funder_award_id":"Andaluc\\u00EDa 2021-2027","funder_id":"https://openalex.org/F4320335322","funder_display_name":"European Regional Development Fund"},{"id":"https://openalex.org/G6621546169","display_name":null,"funder_award_id":"+i 2024/03","funder_id":"https://openalex.org/F4320325313","funder_display_name":"Universidad de C\u00e1diz"},{"id":"https://openalex.org/G7606372943","display_name":null,"funder_award_id":"FPU21/02026","funder_id":"https://openalex.org/F4320315062","funder_display_name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades"}],"funders":[{"id":"https://openalex.org/F3619780700","display_name":"Consejer\u00eda de Universidad, Investigaci\u00f3n e Innovaci\u00f3n, Junta de Andaluc\u00eda","ror":null},{"id":"https://openalex.org/F4320315062","display_name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","ror":null},{"id":"https://openalex.org/F4320319605","display_name":"Consejer\u00eda de Transformaci\u00f3n Econ\u00f3mica, Industria, Conocimiento y Universidades","ror":null},{"id":"https://openalex.org/F4320325313","display_name":"Universidad de C\u00e1diz","ror":"https://ror.org/04mxxkb11"},{"id":"https://openalex.org/F4320335322","display_name":"European Regional Development Fund","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320338080","display_name":"European Social Fund","ror":"https://ror.org/00k4n6c32"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":23,"referenced_works":["https://openalex.org/W2072198393","https://openalex.org/W2169934304","https://openalex.org/W2474835145","https://openalex.org/W2963520355","https://openalex.org/W3034368386","https://openalex.org/W3085139254","https://openalex.org/W3176617251","https://openalex.org/W3204750696","https://openalex.org/W4291653336","https://openalex.org/W4318464561","https://openalex.org/W4377115595","https://openalex.org/W4386736435","https://openalex.org/W4386826409","https://openalex.org/W4388430442","https://openalex.org/W4391855109","https://openalex.org/W4399449628","https://openalex.org/W4401474635","https://openalex.org/W4403983018","https://openalex.org/W4404670736","https://openalex.org/W4404918643","https://openalex.org/W4406214400","https://openalex.org/W4407025681","https://openalex.org/W4412889635"],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"become":[5],"widely":[6],"used":[7],"tools":[8],"across":[9],"domains":[10],"such":[11],"as":[12,67],"education":[13],"and":[14,25,39,72,93,121,138,163,176,204,217,226],"scientific":[15],"research,":[16],"following":[17],"recent":[18],"progress":[19],"in":[20,161,214,219],"computing":[21],"infrastructure,":[22],"data":[23],"availability,":[24],"training":[26],"techniques.":[27],"However,":[28],"their":[29,36,201,205],"increasing":[30],"deployment":[31],"raises":[32],"significant":[33],"concerns":[34],"about":[35],"high":[37],"computational":[38],"energy":[40,134,162,178,215,233],"demands":[41],"at":[42],"inference":[43,61],"time.":[44,164],"This":[45,63],"work":[46],"tackles":[47],"the":[48,129,143,148,152,156,169,183,196,208,237],"problem":[49,71],"from":[50],"a":[51,68,75,89,94,103,241],"software-stack":[52],"perspective,":[53],"presenting":[54],"an":[55],"automatic":[56],"optimization":[57,70,173,243],"pipeline":[58],"for":[59,114,142,155,182],"LLM":[60,247],"engines.":[62],"task":[64],"is":[65],"formulated":[66],"combinatorial":[69],"addressed":[73],"using":[74,110],"cellular":[76],"genetic":[77],"algorithm":[78,87],"that":[79,228],"evolves":[80],"variable-length":[81],"sequences":[82],"of":[83],"code":[84],"optimizations.":[85],"The":[86],"proposes":[88],"novel":[90],"density-guided":[91],"mutation":[92],"longest":[95],"common":[96],"subsequence":[97],"crossover":[98],"operator.":[99],"Experiments":[100],"conducted":[101],"on":[102],"Mac":[104],"Mini":[105],"with":[106,188],"Apple":[107],"Silicon":[108],"M4":[109],"llama.cpp":[111],"demonstrate":[112,223],"that,":[113],"two":[115,184],"Llama":[116],"3.2":[117],"variants":[118],"(with":[119],"1":[120],"3":[122],"billion":[123],"parameters)":[124],"quantized":[125],"to":[126,147,200],"4-bit":[127],"precision,":[128],"optimized":[130,185,197],"engine":[131],"reduces":[132],"total":[133],"consumption":[135,216],"by":[136,140],"13.5%":[137],"runtime":[139],"19.5%":[141],"1-billion-parameter":[144],"model,":[145,239],"compared":[146],"non-optimized":[149,209],"version,":[150],"while":[151],"corresponding":[153],"reductions":[154],"3-billion-parameter":[157],"model":[158],"were":[159,211],"5.3%":[160],"If":[165],"we":[166],"compare":[167],"against":[168],"aggressive":[170],"-O3":[171],"compilation":[172],"flag,":[174],"4.4%":[175],"2.7%":[177],"savings":[179,234],"are":[180],"achieved":[181],"engines,":[186],"respectively,":[187],"comparable":[189],"runtimes.":[190],"It":[191],"was":[192],"also":[193],"studied":[194],"how":[195],"engines":[198],"generalize":[199],"8-bit":[202],"versions,":[203],"improvements":[206],"over":[207],"baseline":[210],"around":[212],"12%":[213],"11.5%":[218],"runtime.":[220],"These":[221],"results":[222],"cross-quantization":[224],"generalization":[225],"confirm":[227],"hardware-aware":[229],"optimizations":[230],"can":[231],"deliver":[232],"without":[235],"modifying":[236],"neural":[238],"offering":[240],"portable":[242],"methodology":[244],"toward":[245],"greener":[246],"deployment.":[248]},"counts_by_year":[],"updated_date":"2026-06-19T15:47:20.252518","created_date":"2026-03-20T00:00:00"}
