Papers › MLPerf Inference Benchmark

MLPerf Inference Benchmark

6 Nov 2019arXiv:1911.02549archive 2025-07-28

Vijay Janapa Reddi, Christine Cheng, David Kanter, Peter Mattson, Guenther Schmuelling, Carole-Jean Wu, Brian Anderson, Maximilien Breughe, Mark Charlebois, William Chou, Ramesh Chukka, Cody Coleman, Sam Davis, Pan Deng, Greg Diamos, Jared Duke, Dave Fick, J. Scott Gardner, Itay Hubara, Sachin Idgunji, Thomas B. Jablin, Jeff Jiao, Tom St. John, Pankaj Kanwar, David Lee, Jeffery Liao, Anton Lokhmotov, Francisco Massa, Peng Meng, Paulius Micikevicius, Colin Osborne, Gennady Pekhimenko, Arun Tejusve Raghunath Rajan, Dilip Sequeira, Ashish Sirasao, Fei Sun, Hanlin Tang, Michael Thomson, Frank Wei, Ephrem Wu, Lingjie Xu, Koichi Yamada, Bing Yu, George Yuan, Aaron Zhong, Peizhao Zhang, Yuchen Zhou

Machine-learning (ML) hardware and software system demand is burgeoning. Driven by ML applications, the number of different ML inference systems has exploded. Over 100 organizations are building ML inference chips, and the systems that incorporate existing models span at least three orders of magnitude in power consumption and five orders of magnitude in performance; they range from embedded devices to data-center solutions. Fueling the hardware are a dozen or more software frameworks and libraries. The myriad combinations of ML hardware and ML software make assessing ML-system performance in an architecture-neutral, representative, and reproducible manner challenging. There is a clear need for industry-wide standard ML benchmarking and evaluation criteria. MLPerf Inference answers that call. In this paper, we present our benchmarking method for evaluating ML inference systems. Driven by more than 30 organizations as well as more than 200 ML engineers and practitioners, MLPerf prescribes a set of rules and best practices to ensure comparability across systems with wildly differing architectures. The first call for submissions garnered more than 600 reproducible inference-performance measurements from 14 organizations, representing over 30 systems that showcase a wide range of capabilities. The submissions attest to the benchmark's flexibility and adaptability.

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 21 code samples harvested from 1 repository linked to this paper: 18 ran (of which 0 constructed an object rather than computing a result; 18 with no instrument failure: 0 honoured, 0 violated, 18 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: community (archive-listed): 21 samples from 1 repository, 18 ran (18 with no instrument failure; 0 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.

gfursin/browser-extension-for-reproducible-research mentioned on GitHubApache-2.0Syntology: not harvested report
gfursin/ck-browser-addon mentioned on GitHubApache-2.0Syntology: not harvested report
gfursin/ck-browser-addon-firefox mentioned on GitHubApache-2.0Syntology: not harvested report
mlcommons/inference mentioned on GitHubpytorchApache-2.0Syntology: 21 samples, 18 ran (0 instrument-failed) · 3 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

21 samples harvested: 18 ran (of which 0 constructed an object rather than computing a result; 18 with no instrument failure: 0 honoured, 0 violated, 18 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

18ran
3unverified

Licence: 0 of the 21 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 mlcommons/inference. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

mlcommons/inference community (archive-listed) 21 samples, 18 ran (0 instrument-failed) · 3 unverified
apply_argmax vision/medical_imaging/3d-unet-kits19/inference_utils.py ran fingerprinted Apache-2.0 (permissive) · 3867e52e00cf507d · report fixture: signature image:ndarray/4/float · returned ndarray · level recorded 2026-09-24
apply_norm_map vision/medical_imaging/3d-unet-kits19/inference_utils.py ran fingerprinted Apache-2.0 (permissive) · 3cea92bceb1a2c9b · report fixture: signature image:Tensor/4/float, norm_map:Tensor/4/float · returned Tensor · level recorded 2026-09-24
calculate_retrieval_metrics e2e-rag/accuracy_eval.py ran fingerprinted Apache-2.0 (permissive) · 49b7d9bb406a744e · report fixture: signature retrieved_urls:list/1/str, expected_urls:list/1/str · returned dict · level recorded 2026-09-24
calculate_retrieval_metrics e2e-rag/evaluation.py ran Apache-2.0 (permissive) · ad3ffba7eb3bd285 · report fixture: signature expected_urls:list/1/str, retrieved_urls:list/1/str, k_values:list/1/int · returned dict · level recorded 2026-09-24
call_judge e2e-rag/accuracy_eval.py ran Apache-2.0 (permissive) · 7a7d20420db85d84 · report fixture: signature question:str/0/str, ground_truth:str/0/str, llm_answer:str/0/str, service_url:str/0/str, model_name:str/0/str, api_key:str/0/str · returned dict · level recorded 2026-09-24
call_judge e2e-rag/evaluate.py ran Apache-2.0 (permissive) · 1ce3b34b77863636 · report fixture: signature session:callable/0/?, service_url:str/0/str, model:str/0/str, question:str/0/str, gold:str/0/str, pred:str/0/str · returned tuple · level recorded 2026-09-24
compute_md5 e2e-rag/datasetup_accuracy_eval.py ran Apache-2.0 (permissive) · 007cfc74e0e0649b · report fixture: signature file_path:str/0/str · returned str · level recorded 2026-09-24
evaluate_results e2e-rag/accuracy_eval.py ran Apache-2.0 (permissive) · 54a637337a4e3cb3 · report fixture: signature results:dict/0/?, dataset_path:str/0/str, num_workers:int/0/int, judge_service_url:str/0/str, judge_model:str/0/str · returned dict · level recorded 2026-09-24
evaluate_retrieval_query e2e-rag/evaluation.py ran Apache-2.0 (permissive) · 305c35ca9e02614f · report fixture: signature rag_db:MockRAGDB/0/?, query:str/0/str, expected_urls:list/1/str, top_k_retriever:int/0/int, top_k_reranking:int/0/int, verbose:bool/0/bool, no_rerank:bool/0/bool, retrieval_strategy:str/0/str, print_results:bool/0/bool, return_results:bool/0/bool · returned dict · level recorded 2026-09-24
extract_wikipedia_links e2e-rag/download_docs.py ran Apache-2.0 (permissive) · 757290e048b0c991 · report fixture: signature item:list/1/str · returned list · level recorded 2026-09-24
fix_malformed_url e2e-rag/download_docs.py ran fingerprinted Apache-2.0 (permissive) · 4ceeb0dd1f0e41e8 · report fixture: signature url:str/0/str · returned str · level recorded 2026-09-24
load_config in 389 papers text_to_video/wan-2.2-t2v-a14b/run_inference.py ran Apache-2.0 (permissive) · 44c6d1dcec31f489 · report fixture: signature config_path:str/0/str · returned dict · level recorded 2026-09-24
load_prompts in 41 papers text_to_video/wan-2.2-t2v-a14b/run_inference.py ran Apache-2.0 (permissive) · 7b3e0f2610f104fb · report fixture: signature dataset_path:str/0/str · returned list · level recorded 2026-09-24
load_results in 82 papers e2e-rag/evaluate.py ran Apache-2.0 (permissive) · a9ede96a0650926d · report fixture: signature path:PosixPath/0/? · returned dict · level recorded 2026-09-24
parse_accuracy_log e2e-rag/datasetup_accuracy_eval.py ran Apache-2.0 (permissive) · 97cde6f95685bb1a · report fixture: signature log_path:str/0/str · returned dict · level recorded 2026-09-24
postprocess in 70 papers
census: 5 implementations from 5 papers, 5 distinct outputs across 2 buckets, one shared input per bucket
lon/network_SUT.py ran Apache-2.0 (permissive) · f5c0874dcb5d50cd · report fixture: signature query:dict/0/? · returned dict · level recorded 2026-09-24
preprocess in 284 papers
census: 11 implementations from 13 papers, 10 distinct outputs across 5 buckets, one shared input per bucket
lon/network_SUT.py ran fingerprinted Apache-2.0 (permissive) · e05733aa3409d7b2 · report fixture: signature query:str/0/str · returned str · level recorded 2026-09-24
validate_database e2e-rag/datasetup_accuracy_eval.py ran fingerprinted Apache-2.0 (permissive) · 7d14cc5656b8c82f · report fixture: signature database_path:str/0/str, retriever_model:str/0/str · returned dict · level recorded 2026-09-24
dnn_model lon/network_SUT.py unverified Apache-2.0 (permissive) · 2d9995b9375a7f1f · report
download_frames_dataset e2e-rag/download_docs.py unverified Apache-2.0 (permissive) · d9fde455e8c45d1d · report

Tasks

Benchmarking

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