{"id":"https://openalex.org/W7204237409","doi":"https://doi.org/10.48550/arxiv.2608.24650","title":"Simthesizer: An Agent-Driven Simulation Framework for LLM Serving Systems","display_name":"Simthesizer: An Agent-Driven Simulation Framework for LLM Serving Systems","publication_year":2026,"publication_date":"2026-08-25","ids":{"openalex":"https://openalex.org/W7204237409","doi":"https://doi.org/10.48550/arxiv.2608.24650"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.24650","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24650","pdf_url":"https://arxiv.org/pdf/2608.24650","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.24650","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114201443","display_name":"Wonung Kim","orcid":"https://orcid.org/0009-0000-1180-0503","observed_orcids":["https://orcid.org/0009-0000-1180-0503"]},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Kim, Wonung","raw_affiliation_strings":["KAIST Daejeon , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST Daejeon , Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100997517","display_name":"Hyunmin Choi","orcid":"https://orcid.org/0009-0002-4116-3896","observed_orcids":["https://orcid.org/0009-0002-4116-3896"]},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Choi, Hyunmin","raw_affiliation_strings":["KAIST Daejeon , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST Daejeon , Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148298822","display_name":"Minsu Kim","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Kim, Minsu","raw_affiliation_strings":["Jongse Park KAIST Daejeon , Republic of Korea","KAIST Daejeon , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Jongse Park KAIST Daejeon , Republic of Korea","institution_ids":["https://openalex.org/I157485424"]},{"raw_affiliation_string":"KAIST Daejeon , Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029070493","display_name":"Jaehong Cho","orcid":"https://orcid.org/0009-0008-8973-5164","observed_orcids":["https://orcid.org/0009-0008-8973-5164"]},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Cho, Jaehong","raw_affiliation_strings":["KAIST Daejeon , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST Daejeon , Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010891394","display_name":"Yeong Wook Kim","orcid":"https://orcid.org/0000-0003-3282-1164","observed_orcids":["https://orcid.org/0000-0003-3282-1164"]},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Kim, Yeongwook","raw_affiliation_strings":["KAIST Daejeon , Republic of Korea"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"KAIST Daejeon , Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5037553165","display_name":"Jongse Park","orcid":"https://orcid.org/0000-0002-6629-449X","observed_orcids":["https://orcid.org/0000-0002-6629-449X"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Park, Jongse","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"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/T12127","display_name":"Software System Performance and Reliability","score":0.2655999958515167,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T12127","display_name":"Software System Performance and Reliability","score":0.2655999958515167,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T11195","display_name":"Simulation Techniques and Applications","score":0.21449999511241913,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10101","display_name":"Cloud Computing and Resource Management","score":0.15729999542236328,"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/system-level-simulation","display_name":"system-level simulation","score":0.7540000081062317},{"id":"https://openalex.org/keywords/ai-coding-agents","display_name":"coding agents","score":0.5950000286102295},{"id":"https://openalex.org/keywords/agentic-workflows","display_name":"agentic workflows","score":0.703000009059906},{"id":"https://openalex.org/keywords/vllm","display_name":"vLLM","score":0.5870000123977661}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6678000092506409},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.6388000249862671},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.5390999913215637},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.4880000054836273},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.44670000672340393},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.4350999891757965},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.39329999685287476},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.3723999857902527},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.34940001368522644},{"id":"https://openalex.org/C201995342","wikidata":"https://www.wikidata.org/wiki/Q682496","display_name":"Systems engineering","level":1,"score":0.34709998965263367},{"id":"https://openalex.org/C113364801","wikidata":"https://www.wikidata.org/wiki/Q26674","display_name":"High fidelity","level":2,"score":0.3411000072956085},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.32019999623298645},{"id":"https://openalex.org/C201203610","wikidata":"https://www.wikidata.org/wiki/Q5157524","display_name":"Computer architecture simulator","level":2,"score":0.2872999906539917},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.26899999380111694},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2630000114440918},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.2621000111103058},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.25459998846054077},{"id":"https://openalex.org/C17500928","wikidata":"https://www.wikidata.org/wiki/Q959968","display_name":"Control system","level":2,"score":0.25060001015663147}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.24650","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24650","pdf_url":"https://arxiv.org/pdf/2608.24650","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.24650","doi":"https://doi.org/10.48550/arxiv.2608.24650","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.24650","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:2608.24650","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24650","pdf_url":"https://arxiv.org/pdf/2608.24650","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":"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/W7204237409.pdf","grobid_xml":"https://content.openalex.org/works/W7204237409.grobid-xml"},"referenced_works_count":4,"referenced_works":["https://openalex.org/W4414939355","https://openalex.org/W4416017756","https://openalex.org/W7150971697","https://openalex.org/W7172436489"],"related_works":["https://openalex.org/W4404848672","https://openalex.org/W7164030167","https://openalex.org/W7162010250","https://openalex.org/W7162458303","https://openalex.org/W7220897122","https://openalex.org/W7160601217","https://openalex.org/W4414688324","https://openalex.org/W4416440248","https://openalex.org/W7160821619","https://openalex.org/W4412886853"],"abstract_inverted_index":{"System-level":[0],"simulation":[1,54],"is":[2],"an":[3,65],"essential":[4],"tool":[5],"for":[6,158,184],"exploring":[7],"the":[8,52,78,106,111,162],"rapidly":[9],"expanding":[10],"design":[11],"space":[12],"of":[13,153],"LLM":[14,27],"serving":[15,28,75,108],"systems,":[16],"where":[17],"real":[18,175],"deployments":[19],"remain":[20],"costly":[21],"and":[22,38,41,77,117,145,166,199,207],"often":[23],"infeasible.":[24],"However,":[25],"modern":[26],"now":[29],"evolves":[30],"faster":[31,201],"than":[32,202],"human-driven":[33],"simulator":[34,95,101,151],"development":[35,71],"can":[36],"track,":[37],"emerging":[39],"workloads":[40],"mechanisms,":[42],"from":[43],"agentic":[44],"workflows":[45],"to":[46,197],"disaggregated":[47],"serving,":[48],"no":[49],"longer":[50],"fit":[51],"monolithic":[53],"pipeline":[55],"that":[56,80,92,103,114],"existing":[57,188],"simulators":[58,79],"assume.":[59],"Each":[60],"new":[61,156],"mechanism":[62],"therefore":[63],"demands":[64],"invasive":[66],"rewrite,":[67],"leaving":[68],"a":[69,90,99,121,130,155,173],"widening":[70],"gap":[72],"between":[73],"deployed":[74],"systems":[76],"model":[81],"them.":[82],"To":[83],"close":[84],"this":[85,140],"gap,":[86],"we":[87],"present":[88],"Simthesizer,":[89],"framework":[91],"realizes":[93,118],"agent-driven":[94],"development.":[96],"Simthesizer":[97,126,171,193],"introduces":[98],"composable":[100],"infrastructure":[102],"uniformly":[104],"expresses":[105],"complete":[107],"workflow,":[109],"including":[110],"control":[112],"decisions":[113],"coordinate":[115],"it,":[116],"it":[119],"as":[120],"unified":[122],"dynamic":[123],"graph":[124],"in":[125],"simulator.":[127],"Synthesizer":[128],"agent,":[129,133],"harnessed":[131],"coding":[132,164],"then":[134],"lowers":[135],"natural-language":[136],"feature":[137],"requests":[138],"onto":[139],"abstraction":[141],"under":[142],"simulator-specific":[143],"guardrails":[144],"fidelity":[146],"validation,":[147],"evolving":[148],"one":[149,157],"shared":[150],"instead":[152],"building":[154],"every":[159],"feature.":[160],"Under":[161],"same":[163],"agent":[165],"harnesses,":[167],"extensions":[168,185],"built":[169,186],"on":[170,187],"follow":[172],"vLLM-based":[174],"system":[176],"with":[177],"2.51%":[178],"average":[179],"throughput":[180],"error,":[181],"versus":[182],"6.03%":[183],"simulators.":[189],"On":[190],"identical":[191],"workloads,":[192],"also":[194],"simulates":[195],"up":[196],"284.96x":[198],"23.19x":[200],"two":[203],"state-of-the-art":[204],"simulators,":[205],"LLMServingSim2.0":[206],"Vidur,":[208],"respectively.":[209]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-27T00:00:00"}