Papers › Fairness in Serving Large Language Models

Fairness in Serving Large Language Models

31 Dec 2023arXiv:2401.00588archive 2025-07-28

Ying Sheng, Shiyi Cao, Dacheng Li, Banghua Zhu, Zhuohan Li, Danyang Zhuo, Joseph E. Gonzalez, Ion Stoica

High-demand LLM inference services (e.g., ChatGPT and BARD) support a wide range of requests from short chat conversations to long document reading. To ensure that all client requests are processed fairly, most major LLM inference services have request rate limits, to ensure that no client can dominate the request queue. However, this rudimentary notion of fairness also results in under-utilization of the resources and poor client experience when there is spare capacity. While there is a rich literature on fair scheduling, serving LLMs presents new challenges due to their unpredictable request lengths and their unique batching characteristics on parallel accelerators. This paper introduces the definition of LLM serving fairness based on a cost function that accounts for the number of input and output tokens processed. To achieve fairness in serving, we propose a novel scheduling algorithm, the Virtual Token Counter (VTC), a fair scheduler based on the continuous batching mechanism. We prove a 2x tight upper bound on the service difference between two backlogged clients, adhering to the requirement of work-conserving. Through extensive experiments, we demonstrate the superior performance of VTC in ensuring fairness, especially in contrast to other baseline methods, which exhibit shortcomings under various conditions. The reproducible code is available at https://github.com/Ying1123/VTC-artifact

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 13 code samples harvested from 2 repositories linked to this paper: 9 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 6 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 6 samples from 1 repository, 4 ran (1 with no instrument failure; 3 where Syntology's instrument failed); community (archive-listed): 7 samples from 1 repository, 5 ran (2 with no instrument failure; 3 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

ying1123/vtc-artifact officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 6 samples, 4 ran (3 instrument-failed) · 2 unverified report
s-lora/s-lora mentioned on GitHubpytorchApache-2.0Syntology: 7 samples, 5 ran (3 instrument-failed) · 2 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

13 samples harvested: 9 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 6 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

6ran · our draft was wrong
3ran
4unverified

Licence: 0 of the 13 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

ying1123/vtc-artifact official repository 6 samples, 4 ran (3 instrument-failed) · 2 unverified
dummy_prompt benchmarks/trace.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · cf380cfb6bd64b21 · report fixture: signature prompt_len:int/0/int · returned str · level recorded 2026-09-24
generate_requests benchmarks/trace.py ran Apache-2.0 (permissive) · 2bf66d26ba8a4548 · report fixture: signature num_adapters:int/0/int, alpha:float/0/float, req_rate:int/0/int, cv:float/0/float, duration:int/0/int, input_range:tuple/1/int, output_range:tuple/1/int, adapter_dirs:list/1/tuple, seed:int/0/int · returned list · level recorded 2026-09-24
to_dict in 13 papers fair_bench/exp_suite.py ran · our draft was wrong Apache-2.0 (permissive) · ecdd469e9c7b7443 · report fixture: signature config:BenchmarkConfig/1/int · returned dict · level recorded 2026-09-24
to_tuple in 6 papers fair_bench/exp_suite.py ran · our draft was wrong Apache-2.0 (permissive) · 98027d2de0638849 · report fixture: signature config:dict/0/? · returned tuple · level recorded 2026-09-24
get_all_suites fair_bench/exp_suite.py unverified Apache-2.0 (permissive) · 3b155d666ec2f922 · report
get_real_requests benchmarks/trace.py unverified Apache-2.0 (permissive) · ceccd6265c698888 · report
s-lora/s-lora community (archive-listed) 7 samples, 5 ran (3 instrument-failed) · 2 unverified
benchmark in 19 papers benchmarks/run_exp.py ran Apache-2.0 (permissive) · 6e238d2d144eed9b · report fixture: signature backend:str/0/str, server:str/0/str, input_requests:list/1/Request, debug:bool/0/bool · returned coroutine · level recorded 2026-09-24
get_lora_config_json slora/mprophet/lora_config.py ran · our draft was wrong Apache-2.0 (permissive) · 5870f80ad0e8422b · report fixture: signature name:str/0/str · returned dict · level recorded 2026-09-24
send_request benchmarks/run_exp.py ran Apache-2.0 (permissive) · 3719da4f729ce737 · report fixture: signature backend:str/0/str, server:str/0/str, req_id:str/0/str, model_dir:str/0/str, adapter_dir:str/0/str, prompt:str/0/str, prompt_len:int/0/int, output_len:int/0/int, debug:bool/0/bool · returned coroutine · level recorded 2026-09-24
to_dict in 13 papers benchmarks/exp_suite.py ran · our draft was wrong Apache-2.0 (permissive) · a027792a13b85ec3 · report fixture: signature config:BenchmarkConfig/1/int · returned dict · level recorded 2026-09-24
to_tuple in 6 papers benchmarks/exp_suite.py ran · our draft was wrong Apache-2.0 (permissive) · 1fb624c4bbfa4098 · report fixture: signature config:dict/0/? · returned tuple · level recorded 2026-09-24
get_all_suites benchmarks/exp_suite.py unverified Apache-2.0 (permissive) · 03e5db79fc455665 · report
get_peak_mem benchmarks/run_exp.py unverified Apache-2.0 (permissive) · 72eaf41bbc94d98b · report

Tasks

FairnessScheduling

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections