Papers › LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale

LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale

15 Aug 2022arXiv:2208.07339archive 2025-07-28

Tim Dettmers, Mike Lewis, Younes Belkada, Luke Zettlemoyer

Large language models have been widely adopted but require significant GPU memory for inference. We develop a procedure for Int8 matrix multiplication for feed-forward and attention projection layers in transformers, which cut the memory needed for inference by half while retaining full precision performance. With our method, a 175B parameter 16/32-bit checkpoint can be loaded, converted to Int8, and used immediately without performance degradation. This is made possible by understanding and working around properties of highly systematic emergent features in transformer language models that dominate attention and transformer predictive performance. To cope with these features, we develop a two-part quantization procedure, LLM.int8(). We first use vector-wise quantization with separate normalization constants for each inner product in the matrix multiplication, to quantize most of the features. However, for the emergent outliers, we also include a new mixed-precision decomposition scheme, which isolates the outlier feature dimensions into a 16-bit matrix multiplication while still more than 99.9% of values are multiplied in 8-bit. Using LLM.int8(), we show empirically it is possible to perform inference in LLMs with up to 175B parameters without any performance degradation. This result makes such models much more accessible, for example making it possible to use OPT-175B/BLOOM on a single server with consumer GPUs. We open-source our software.

PaperPDFCodeCode Syntology ranMethod as code

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 5 code samples harvested from 2 repositories linked to this paper: 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. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed); community (archive-listed): 4 samples from 1 repository, 4 ran (4 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.

No sample is linked to this paper from 2 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-24 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

timdettmers/bitsandbytes officialmentioned in papermentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified report
alextmallen/adaptive-retrieval mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to When Not to Trust Language Models: Investigating Effectiveness of Parametric and... (arXiv:2212.10511; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
huggingface/transformers-bloom-inference mentioned on GitHubpytorchApache-2.0Syntology: 4 samples, 4 ran (0 instrument-failed) · 0 unverified report
kohjingyu/fromage mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Grounding Language Models to Images for Multimodal Inputs and Outputs (arXiv:2301.13823; that paper's own run record: 6 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; 2 where Syntology's instrument failed) · 2 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

5 samples harvested: 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. “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.

4ran
1unverified

Licence: 2 of the 5 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. 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.

timdettmers/bitsandbytes official repository 1 sample, 0 ran · 1 unverified
int8_linear_matmul bitsandbytes/functional.py unverified MIT (permissive) · 646f7cd7192b9ea8 · report
huggingface/transformers-bloom-inference community (archive-listed) 4 samples, 4 ran (0 instrument-failed) · 0 unverified
get_benchmark_results inference_server/benchmark.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 5f7de07e75040659 · report fixture: signature benchmark_time:float/0/float, initialization_time:float/0/float, total_new_tokens_generated:int/0/int, batch_size:int/0/int, cycles:int/0/int · returned str · level recorded 2026-09-24
get_hf_model_class inference_server/models/model.py ran Apache-2.0 (permissive) · 9442182e69129d23 · report fixture: signature model_class:str/0/str · returned type · level recorded 2026-09-25
get_model_path in 22 papers inference_server/models/ds_inference.py ran fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 23648dfbe89d08f8 · report fixture: signature model_name:str/0/str · returned str · level recorded 2026-09-24
load_tokenizer in 66 papers inference_server/models/model.py ran Apache-2.0 (permissive) · 835634e64b6b277d · report fixture: signature model_name:str/0/str · returned Mock · level recorded 2026-09-25

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for LLM.int8() (graph method: LLM.int8()). 1 brick; individually on a fixture: 0 ran · 1 unverified. The assembly has not been run end-to-end.

  1. int8_linear_matmul entry · level 0 · unverified · 646f7cd7192b9ea8

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2208.07339") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Language ModellingLinguistic AcceptabilityNatural Language InferenceQuantizationSemantic Textual SimilaritySentiment Analysis

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Language Modelling C4 Zeropoint LLM.int8 13B (vector-wise + decomp) Perplexity 12.45 #2 of 9 Archive leaderboard report
Language Modelling C4 LLM.float32 6.7B Perplexity 13.3 #5 of 9 Archive leaderboard report
Language Modelling C4 LLM.float32 2.7B Perplexity 14.43 #6 of 9 Archive leaderboard report
Language Modelling C4 LLM.float32 1.3B Perplexity 15.91 #9 of 9 Archive leaderboard report
Linguistic Acceptability CoLA RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned) Accuracy 68.6% #17 of 43 Archive leaderboard report
Natural Language Inference MultiNLI RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned) Matched 90.2 #11 of 67 Archive leaderboard report
Natural Language Inference QNLI RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned) Accuracy 94.7% #13 of 43 Archive leaderboard report
Natural Language Inference RTE RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned) Accuracy 85.4% #25 of 90 Archive leaderboard report
Semantic Textual Similarity MRPC RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned) Accuracy 91.0% #7 of 45 Archive leaderboard report
Semantic Textual Similarity STS Benchmark RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned) Pearson Correlation 0.919 #9 of 66 Archive leaderboard report
Sentiment Analysis SST-2 Binary classification RoBERTa-large 355M (MLP quantized vector-wise, fine-tuned) Accuracy 96.4 #16 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.

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