Papers › Demystifying the Compression of Mixture-of-Experts Through a Unified Framework

Demystifying the Compression of Mixture-of-Experts Through a Unified Framework

4 Jun 2024arXiv:2406.02500archive 2025-07-28

Shwai He, Daize Dong, Liang Ding, Ang Li

Scaling large language models has revolutionized the performance across diverse domains, yet the continual growth in model size poses significant challenges for real-world deployment. The Mixture of Experts (MoE) approach addresses this by dynamically selecting and activating only a subset of experts, significantly reducing computational costs while maintaining high performance. However, MoE introduces potential redundancy (e.g., parameters) and extra costs (e.g., communication overhead). Despite numerous compression techniques developed for mitigating the redundancy in dense models, the compression of MoE remains under-explored. We first bridge this gap with a cutting-edge unified framework that not only seamlessly integrates mainstream compression methods but also helps systematically understand MoE compression. This framework approaches compression from two perspectives: Expert Slimming which compresses individual experts and Expert Trimming which removes structured modules. Within this framework, we explore the optimization space unexplored by existing methods,and further introduce aggressive Expert Trimming techniques, i.e., Layer Drop and Block Drop, to eliminate redundancy at larger scales. Based on these insights,we present a comprehensive recipe to guide practitioners in compressing MoE effectively. Extensive experimental results demonstrate the effectiveness of the compression methods under our framework and the proposed recipe, achieving a 6.05x speedup and only 20.0GB memory usage while maintaining over 92% of performance on Mixtral-8x7B. Code is released at https://github.com/DaizeDong/Unified-MoE-Compression.

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 harvested from 1 repository linked to this paper: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 6 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: 11 samples from 1 repository, 5 ran (4 with no instrument failure; 1 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.

daizedong/unified-moe-compression officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 11 samples, 5 ran (1 instrument-failed) · 6 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

11 samples harvested: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
1ran · our draft was wrong
3ran
6unverified

Licence: 0 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 daizedong/unified-moe-compression. “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.

daizedong/unified-moe-compression official repository 11 samples, 5 ran (1 instrument-failed) · 6 unverified
check_sparsity_from_state_dict src/llmtuner/compression/prune/utils.py ran · honoured contract Apache-2.0 (permissive) · 7b483277d903ceb1 · report fixture: signature state_dict:dict/0/? · returned float · level recorded 2026-09-24
fill_missing_values_for_non_moe_layers src/llmtuner/compression/prune/expert_drop.py ran Apache-2.0 (permissive) · f7e6b2633db22c17 · report fixture: signature values:list/1/int, moe_layer_indices:list/1/int, num_layers:int/0/int · returned list · level recorded 2026-09-24
find_modules in 6 papers src/llmtuner/compression/prune/utils.py ran Apache-2.0 (permissive) · 6031b360d9fd7f04 · report fixture: signature module:callable/0/?, layers:list/1/callable, name:str/0/str · returned dict · level recorded 2026-09-24
find_moe_expert_linears src/llmtuner/compression/prune/utils.py ran Apache-2.0 (permissive) · d857212c7c79f1fe · report fixture: signature module:callable/0/?, exclude_names:NoneType/0/? · returned dict · level recorded 2026-09-24
load_json in 344 papers src/llmtuner/compression/prune/io.py ran · our draft was wrong Apache-2.0 (permissive) · 8db0c1b19aba233f · report fixture: signature file_path:str/0/str · returned dict · level recorded 2026-09-24
apply_moe_weights AutoAWQ/awq/modules/fused/moe.py unverified Apache-2.0 (permissive) · 858343dca0bec6c6 · report
fused_topk AutoAWQ/awq/modules/fused/moe.py unverified Apache-2.0 (permissive) · a1324de2a7fd6559 · report
moe_align_block_size AutoAWQ/awq/modules/fused/moe.py unverified Apache-2.0 (permissive) · 68b3f990b0ec61d6 · report
prune_magnitude src/llmtuner/compression/prune/prune.py unverified Apache-2.0 (permissive) · bdface3275751bca · report
prune_sparsegpt src/llmtuner/compression/prune/prune.py unverified Apache-2.0 (permissive) · 3c36697395392b37 · report
prune_wanda src/llmtuner/compression/prune/prune.py unverified Apache-2.0 (permissive) · 477c9e48c3c1bac2 · report

Tasks

Mixture-of-Experts

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

MoE

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