SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.14808 · ACL · 2024

RelayAttention for Efficient Large Language Model Serving with Long System Prompts

Xinjiang Wang, Rynson Lau, Wayne Zhang, Lei Zhu, Cache Op

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2402.14808.json · badge: /api/ran/2402.14808.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
rayleizhu/vllm-ra canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
FunctionStatusWhere it lives
get_lock Ran
ran; our contract draft was wrong, not the code
rayleizhu/vllm-ra/vllm/model_executor/weight_utils.py
code served (permissive licence) · get_code("9f09d0beacdd35d1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned UnixFileLock. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import os
import tempfile

# The function takes model_name_or_path (str) and optional cache_dir (str)
# It uses these to construct a lock file path
# model_name_or_path: a string that will have '/' replaced with '-'
# cache_dir: optional directory path, defaults to '/tmp'

ARGS = {
    'model_name_or_path': 'bert-base-uncased',
    'cache_dir': None
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
filelock, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type UnixFileLock — a UnixFileLock output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a UnixFileLock output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
init_logger Ran
ran; our contract draft was wrong, not the code
rayleizhu/vllm-ra/vllm/logger.py
code served (permissive licence) · get_code("13adeb89dc2bcd5c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Logger. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import logging

# Initialize the default handler that init_logger expects
_default_handler = logging.StreamHandler()
_default_handler.setLevel(logging.DEBUG)

ARGS = {
    "name": "test_logger"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
prepare_hf_model_weights Ran
ran; our contract draft was wrong, not the code
rayleizhu/vllm-ra/vllm/model_executor/weight_utils.py
code served (permissive licence) · get_code("8be17b096145392b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
import torch

# Create a temporary directory to act as a local model repository
temp_dir = tempfile.mkdtemp()

# Create a mock model weights file in the temporary directory
mock_weights_file = os.path.join(temp_dir, "model.safetensors")
with open(mock_weights_file, 'w') as f:
    f.write("mock weights content")

# Define arguments for prepare_hf_model_weights
ARGS = {
    'model_name_or_path': temp_dir,  # Use local directory to avoid network calls
    'cache_dir': None,
    'load_format': 'auto',
    'fall_back_to_pt': True,
    'revision': None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
filelock, huggingface_hub, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
get_lock 3 other papers
1 dated earlier, 2 later
2309.06180 (2023-09)
Efficient Memory Management for Large Language Model Serving with PagedAttention
2403.02310 (2024-03)
Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve
2501.14808 (2025-01)
HyGen: Efficient LLM Serving via Elastic Online-Offline Request Co-location
init_logger 1 other paper
1 dated earlier
2309.06180 (2023-09)
Efficient Memory Management for Large Language Model Serving with PagedAttention

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

A practical large language model (LLM) service may involve a long system prompt, which specifies the instructions, examples, and knowledge documents of the task and is reused across requests. However, the long system prompt causes throughput/latency bottlenecks as the cost of generating the next token grows w.r.t. the sequence length. This paper aims to improve the efficiency of LLM services that involve long system prompts. Our key observation is that handling these system prompts requires heavily redundant memory accesses in existing causal attention computation algorithms. Specifically, for batched requests, the cached hidden states (i.e., key-value pairs) of system prompts are transferred from off-chip DRAM to on-chip SRAM multiple times, each corresponding to an individual request. To eliminate such a redundancy, we propose RelayAttention, an attention algorithm that allows reading these hidden states from DRAM exactly once for a batch of input tokens. RelayAttention is a free lunch: it maintains the generation quality while requiring no model retraining, as it is based on a mathematical reformulation of causal attention. We have observed significant performance improvements to a productionlevel system, vLLM, through integration with RelayAttention. The improvements are even more profound with longer system prompts.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2402.14808")
get_code_for_paper("2402.14808")
have("2402.14808")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2402.14808.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2402.14808.svg)](https://syntology.ai/paper/2402.14808)

Connect an agent — have() is free.