Papers › QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks

QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks

6 Feb 2024arXiv:2402.04396archive 2025-07-28

Albert Tseng, Jerry Chee, Qingyao Sun, Volodymyr Kuleshov, Christopher De Sa

Post-training quantization (PTQ) reduces the memory footprint of LLMs by quantizing their weights to low-precision. In this work, we introduce QuIP#, a weight-only PTQ method that achieves state-of-the-art results in extreme compression regimes (≤ 4 bits per weight) using three novel techniques. First, QuIP# improves QuIP's (Chee et al., 2023) incoherence processing by using the randomized Hadamard transform, which is faster and has better theoretical properties. Second, QuIP# uses vector quantization to take advantage of the ball-shaped sub-Gaussian distribution that incoherent weights possess: specifically, we introduce a set of hardware-efficient codebooks based on the highly symmetric E₈ lattice, which achieves the optimal 8-dimension unit ball packing. Third, QuIP# uses fine-tuning to improve fidelity to the original model. Our experiments show that QuIP# outperforms existing PTQ methods, enables new behaviors in PTQ scaling, and supports fast inference. Our code can be found at https://github.com/Cornell-RelaxML/quip-sharp.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2402.04396")

Code

Syntology Ran 9 of 11 code samples harvested from 1 repository linked to this paper; 2 have no recorded run. Of those that ran: 1 ran · violated contract; 3 ran · our draft was wrong; 3 ran · fixture could not drive it; 2 ran with no contract checked.

By repository: official repository: 11 samples from 1 repository, 9 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

cornell-relaxml/quip-sharp officialmentioned in papermentioned on GitHubpytorchGPL-3.0 report
goodevening13/aquakv mentioned on GitHubpytorchApache-2.0 report
pilancilab/caldera mentioned on GitHubpytorch 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; 9 ran; 0 honoured the contract we drafted; 2 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · violated contract
3ran · our draft was wrong
3ran · fixture could not drive it
2ran
2unverified

Licence: 11 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 Cornell-RelaxML/quip-sharp. “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. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (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 from a repository marked official are listed first. 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 four 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; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). 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.

logits_to_probs Cornell-RelaxML/quip-sharp/eval/interactive_gen.py official repository ran · fixture could not drive it fingerprinted GPL-3.0 (copyleft) · pointer only · 357adcbb4430f3ab · report
repeat_kv Cornell-RelaxML/quip-sharp/model/llama.py official repository ran · fixture could not drive it fingerprinted GPL-3.0 (copyleft) · pointer only · 30d7eec482ebf6b1 · report
apply_rotary_pos_emb Cornell-RelaxML/quip-sharp/model/mistral.py official repository ran · fixture could not drive it GPL-3.0 (copyleft) · pointer only · d61c483a3c2b3156 · report
apply_rotary_pos_emb Cornell-RelaxML/quip-sharp/model/llama.py official repository ran · our draft was wrong GPL-3.0 (copyleft) · pointer only · bac65c3dafaec040 · report
check_exist cornell-relaxml/quip-sharp/quantize_llama/quantize_finetune_llama.py official repository ran · violated contract GPL-3.0 (copyleft) · pointer only · 6ab265cc90f28f05 · report
get_emb Cornell-RelaxML/quip-sharp/quantize_llama/finetune_e2e_llama.py official repository ran GPL-3.0 (copyleft) · pointer only · 71304b57668d33e4 · report
multinomial_sample_one_no_sync Cornell-RelaxML/quip-sharp/eval/interactive_gen.py official repository ran · our draft was wrong fingerprinted GPL-3.0 (copyleft) · pointer only · 0b5f90e8856ad7af · report
rotate_half Cornell-RelaxML/quip-sharp/model/llama.py official repository ran · our draft was wrong fingerprinted GPL-3.0 (copyleft) · pointer only · b99eea6376d1e212 · report
sample Cornell-RelaxML/quip-sharp/eval/interactive_gen.py official repository ran GPL-3.0 (copyleft) · pointer only · 2f36db2b5f31a259 · report
get_full_grid Cornell-RelaxML/quip-sharp/lib/codebook/latticee8_padded12.py official repository unverified GPL-3.0 (copyleft) · pointer only · a723f0dd343b0d27 · report
llama_arg_fn Cornell-RelaxML/quip-sharp/quantize_llama/finetune_e2e_llama.py official repository unverified GPL-3.0 (copyleft) · pointer only · cda0c2641d73d740 · report

Tasks

Quantization

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

SET

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