Papers › FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving

FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving

2 Jan 2025arXiv:2501.01005archive 2025-07-28

Zihao Ye, Lequn Chen, Ruihang Lai, Wuwei Lin, Yineng Zhang, Stephanie Wang, Tianqi Chen, Baris Kasikci, Vinod Grover, Arvind Krishnamurthy, Luis Ceze

Transformers, driven by attention mechanisms, form the foundation of large language models (LLMs). As these models scale up, efficient GPU attention kernels become essential for high-throughput and low-latency inference. Diverse LLM applications demand flexible and high-performance attention solutions. We present FlashInfer: a customizable and efficient attention engine for LLM serving. FlashInfer tackles KV-cache storage heterogeneity using block-sparse format and composable formats to optimize memory access and reduce redundancy. It also offers a customizable attention template, enabling adaptation to various settings through Just-In-Time (JIT) compilation. Additionally, FlashInfer's load-balanced scheduling algorithm adjusts to dynamism of user requests while maintaining compatibility with CUDAGraph which requires static configuration. FlashInfer have been integrated into leading LLM serving frameworks like SGLang, vLLM and MLC-Engine. Comprehensive kernel-level and end-to-end evaluations demonstrate FlashInfer's ability to significantly boost kernel performance across diverse inference scenarios: compared to state-of-the-art LLM serving solutions, FlashInfer achieve 29-69% inter-token-latency reduction compared to compiler backends for LLM serving benchmark, 28-30% latency reduction for long-context inference, and 13-17% speedup for LLM serving with parallel generation.

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 16 code samples harvested from 1 repository linked to this paper: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 11 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 16 samples from 1 repository, 5 ran (5 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 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.

flashinfer-ai/flashinfer officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 16 samples, 5 ran (0 instrument-failed) · 11 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

16 samples harvested: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 11 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.

5ran
11unverified

Licence: 0 of the 16 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 flashinfer-ai/flashinfer. “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.

flashinfer-ai/flashinfer official repository 16 samples, 5 ran (0 instrument-failed) · 11 unverified
min_block_table_width flashinfer/attn_scores/attn_scores.py ran fingerprinted Apache-2.0 (permissive) · 42cb35beca6e09b0 · report fixture: signature seq_len:int/0/int, block_size:int/0/int · returned int · level recorded 2026-09-24
normalize_cuda_architecture flashinfer-jit-cache-provider/package_config.py ran Apache-2.0 (permissive) · 97b676335b993413 · report fixture: signature architecture:str/0/str · returned tuple · level recorded 2026-09-24
padded_seq_len flashinfer/attn_scores/attn_scores.py ran fingerprinted Apache-2.0 (permissive) · 96eca989be463f73 · report fixture: signature max_seq_len:int/0/int · returned int · level recorded 2026-09-24
parse_bool in 9 papers flashinfer/aot.py ran Apache-2.0 (permissive) · c6baeb0a93584501 · report fixture: signature s:str/0/str · returned bool · level recorded 2026-09-25
parse_head_dim flashinfer/aot.py ran Apache-2.0 (permissive) · 3ea5257687b4a413 · report fixture: signature head_dim:str/0/str · returned tuple · level recorded 2026-09-24
build_editable flashinfer-cubin/build_backend.py unverified Apache-2.0 (permissive) · 44a0c38fcf791070 · report
build_wheel flashinfer-cubin/build_backend.py unverified Apache-2.0 (permissive) · 8b66f8de9a6161fa · report
flush_graph_dumps flashinfer/api_logging.py unverified Apache-2.0 (permissive) · 4d8db70f7a999b76 · report
get_act_and_mul_module flashinfer/activation.py unverified Apache-2.0 (permissive) · 3abbd9a8cd490c7a · report
get_available_cubin_files flashinfer/artifacts.py unverified Apache-2.0 (permissive) · 6438bbe2fa20b253 · report
get_available_header_files flashinfer/artifacts.py unverified Apache-2.0 (permissive) · 7de81a2e3fa334d6 · report
get_checksums flashinfer/artifacts.py unverified Apache-2.0 (permissive) · 67f1e79951ba168d · report
prepare_metadata_for_build_wheel flashinfer-jit-cache-provider/build_backend.py unverified Apache-2.0 (permissive) · 21affb9cd865c03b · report
prepare_metadata_for_build_wheel flashinfer-jit-cache/build_backend.py unverified Apache-2.0 (permissive) · 80e22b335608272d · report
replay_from_dump flashinfer/api_logging.py unverified Apache-2.0 (permissive) · 85b3d4c658379be0 · report
replay_sequence flashinfer/api_logging.py unverified Apache-2.0 (permissive) · 3671383ef718172c · report

Tasks

Scheduling

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AttentionSoftmax

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