Papers › Q8BERT: Quantized 8Bit BERT

Q8BERT: Quantized 8Bit BERT

14 Oct 2019arXiv:1910.06188archive 2025-07-28

Ofir Zafrir, Guy Boudoukh, Peter Izsak, Moshe Wasserblat

Recently, pre-trained Transformer based language models such as BERT and GPT, have shown great improvement in many Natural Language Processing (NLP) tasks. However, these models contain a large amount of parameters. The emergence of even larger and more accurate models such as GPT2 and Megatron, suggest a trend of large pre-trained Transformer models. However, using these large models in production environments is a complex task requiring a large amount of compute, memory and power resources. In this work we show how to perform quantization-aware training during the fine-tuning phase of BERT in order to compress BERT by 4× with minimal accuracy loss. Furthermore, the produced quantized model can accelerate inference speed if it is optimized for 8bit Integer supporting hardware.

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 11 code samples: 10 from 3 repositories linked to this paper, 1 identical to code harvested from another paper's repository. Across all 11: 10 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 2 honoured, 0 violated, 7 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 5 samples from 1 repository, 5 ran (5 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 5 samples from 2 repositories, 4 ran (4 with no instrument failure; 0 where Syntology's instrument failed); 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 10 samples that ran has no recorded date.

intellabs/model-compression-research-package officialmentioned on GitHubpytorchnot reachable when probed 2026-09-17 — repositories for recent papers often appear after camera-readySyntology: not harvested report
NervanaSystems/nlp-architect officialtfApache-2.0Syntology: 5 samples, 5 ran (0 instrument-failed) · 0 unverified report
huggingface/block_movement_pruning mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
iabd/QuantizedNMT mentioned on GitHubpytorchApache-2.0Syntology: 3 samples, 2 ran (0 instrument-failed) · 1 unverified report
mindspore-ai/models mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdaBin: Improving Binary Neural Networks with Adaptive Binary Sets (arXiv:2208.08084; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

11 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 9 with no instrument failure: 2 honoured, 0 violated, 7 with no contract checked; 1 where Syntology's instrument failed) · 1 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.

2ran · honoured contract
1ran · our draft was wrong
7ran
1unverified

Licence: 3 of the 11 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 3 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. 1 sample is listed last, in a group without a repository. “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.

NervanaSystems/nlp-architect official repository 5 samples, 5 ran (0 instrument-failed) · 0 unverified
f1 in 57 papers nlp_architect/models/np_semantic_segmentation.py ran Apache-2.0 (permissive) · 6a2e2df3980634be · report fixture: signature y_true:EagerTensor/2/float, y_pred:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-25
find_correct_index nlp_architect/common/core_nlp_doc.py ran Apache-2.0 (permissive) · 17b3dba51524b6d1 · report fixture: signature orig_gov:dict/0/?, merged_punct_sentence:list/1/dict · returned int · level recorded 2026-09-25
merge_punctuation nlp_architect/common/core_nlp_doc.py ran Apache-2.0 (permissive) · e9f89451c8a8ee18 · report fixture: signature sentence:list/1/dict · returned list · level recorded 2026-09-25
precision_score nlp_architect/models/np_semantic_segmentation.py ran Apache-2.0 (permissive) · 8891829bac9ad9a6 · report fixture: signature y_true:EagerTensor/2/float, y_pred:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-25
recall_score nlp_architect/models/np_semantic_segmentation.py ran Apache-2.0 (permissive) · 63b63c3ae6452ee1 · report fixture: signature y_true:EagerTensor/2/float, y_pred:EagerTensor/2/float · returned EagerTensor · level recorded 2026-09-25
huggingface/block_movement_pruning community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
regularization in 12 papers block_movement_pruning/masked_run_squad.py ran · honoured contract no licence file found · pointer only · 1e24f40610cc5f59 · report fixture: signature model:callable/0/?, mode:str/0/str · returned Tensor · level recorded 2026-09-22
schedule_threshold block_movement_pruning/masked_run_squad.py ran · honoured contract fingerprinted no licence file found · pointer only · 484979ea1bdecce9 · report fixture: signature step:int/0/int, total_step:int/0/int, warmup_steps:int/0/int, initial_threshold:float/0/float, final_threshold:float/0/float, initial_warmup:int/0/int, final_warmup:int/0/int, final_lambda:float/0/float, initial_ampere_temperature:float/0/float, final_ampere_temperature:float/0/float, initial_shuffling_temperature:float/0/float, final_shuffling_temperature:float/0/float · returned tuple · level recorded 2026-09-22
iabd/QuantizedNMT community (archive-listed) 3 samples, 2 ran (0 instrument-failed) · 1 unverified
get_std_opt in 6 papers src/noamOpt.py ran Apache-2.0 (permissive) · b08eb2ca2806310a · report fixture: signature model:callable/0/? · returned NoamOpt · level recorded 2026-09-24
selfAttention src/attention.py ran Apache-2.0 (permissive) · 2f6f127ef7168a65 · report fixture: signature q:Tensor/3/float, k:Tensor/3/float, v:Tensor/3/float, mask:Tensor/3/float, dropout:callable/0/? · returned tuple · level recorded 2026-09-24
batchSizeFn Trace.py unverified Apache-2.0 (permissive) · e0a328541873813a · report
Repository not located 1 sample, 1 ran (1 instrument-failed) · 0 unverified
1 sample is identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
to_list in 62 papers repository not located ran · our draft was wrong fingerprinted licence of this copy not recorded · 9df40357afea56cc · report fixture: signature tensor:Tensor/2/float · returned list · level recorded, date not recorded

Tasks

Linguistic AcceptabilityNatural Language InferenceQuantizationSemantic Textual SimilaritySentiment Analysis

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Linguistic Acceptability CoLA Q8BERT (Zafrir et al., 2019) Accuracy 65.0 #24 of 43 Archive leaderboard report
Natural Language Inference MultiNLI Q8BERT (Zafrir et al., 2019) Matched 85.6 #29 of 67 Archive leaderboard report
Natural Language Inference QNLI Q8BERT (Zafrir et al., 2019) Accuracy 93.0 #23 of 43 Archive leaderboard report
Natural Language Inference RTE Q8BERT (Zafrir et al., 2019) Accuracy 84.8 #27 of 90 Archive leaderboard report
Semantic Textual Similarity MRPC Q8BERT (Zafrir et al., 2019) Accuracy 89.7 #17 of 45 Archive leaderboard report
Semantic Textual Similarity STS Benchmark Q8BERT (Zafrir et al., 2019) Pearson Correlation 0.911 #14 of 66 Archive leaderboard report
Sentiment Analysis SST-2 Binary classification Q8BERT (Zafrir et al., 2019) Accuracy 94.7 #32 of 87 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionAttention DropoutBERTBPECosine AnnealingDense ConnectionsDiscriminative Fine-TuningDropoutGPTLabel SmoothingLayer NormalizationLinear LayerLinear Warmup With Cosine AnnealingLinear Warmup With Linear DecayMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSPEEDSoftmaxTransformerWeight DecayWordPiece

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