SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.01027 · 2026

SFMP: Fine-Grained, Hardware-Friendly and Search-Free Mixed-Precision Quantization for Large Language Models

Xin Nie, Liang Dong, Haicheng Zhang, Beining Feng, Jinhong Weng, Guiling Sun

arXiv · PDF · Open in the Atlas

Code that ran

9 of 12 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; 9 failures of our instrument; 3 no recorded run), measured 2026-11-22; 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/2602.01027.json · badge: /api/ran/2602.01027.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
Nkniexin/SFMP canonical (classifier label)
classifier basis: census
9 ran · 3 no recorded run
FunctionStatusWhere it lives
block_replace Ran
ran; our contract draft was wrong, not the code
Nkniexin/SFMP/AWQ/skip_llama.py
pointer only (licence: NONE) · get_code("a7c67560fd174205")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

create_logger Ran
ran; our contract draft was wrong, not the code
Nkniexin/SFMP/sfmp/utils.py
pointer only (licence: NONE) · get_code("87877d0c60858bff")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_alpaca_dataset Ran
ran; our contract draft was wrong, not the code
Nkniexin/SFMP/EfficientQAT/datautils_e2e.py
pointer only (licence: NONE) · get_code("b5445674ab17410f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

find_layers Ran
ran; our contract draft was wrong, not the code
Nkniexin/SFMP/GPTQ/modelutils.py
pointer only (licence: NONE); the stored copy get_code serves, from squeezeailab/kvquant, is pointer only (licence: NONE) · get_code("a9e7f2cdf016b88b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_average_bit Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Nkniexin/SFMP/sfmp/bitallocation.py
pointer only (licence: NONE) · get_code("e8a53ceca8088e39")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_block_importance Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Nkniexin/SFMP/sfmp/bitallocation.py
pointer only (licence: NONE) · get_code("d88a9761f705c768")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_named_linears Ran
ran; our contract draft was wrong, not the code
Nkniexin/SFMP/sfmp/utils.py
pointer only (licence: NONE) · get_code("e39a65a9a0dc2400")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

pack_bits_to_int32 Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Nkniexin/SFMP/sfmp/BCQLinear.py
pointer only (licence: NONE) · get_code("3d8e69bac57acb96")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

pack_int4 Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
Nkniexin/SFMP/sfmp/utils.py
pointer only (licence: NONE) · get_code("92b6798bd92fcc02")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

assign_importance_levels Not yet run
unverified (harvested, no recorded run)
Nkniexin/SFMP/AWQ/awq.py
pointer only (licence: NONE) · get_code("5316cfb21763e6f4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

bit_alloc_block_wise Not yet run
unverified (harvested, no recorded run)
Nkniexin/SFMP/AWQ/awq.py
pointer only (licence: NONE) · get_code("be6adeb60e252a21")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

direct_block Not yet run
unverified (harvested, no recorded run)
Nkniexin/SFMP/AWQ/awq.py
pointer only (licence: NONE) · get_code("5aaf7ec4c1c74b1d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Nkniexin/SFMP when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 0 later) and 9 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
find_layers 27 other papers
26 dated earlier, 1 undated
2203.13474 (2022-03)
CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis
2023.emnlp-main.892 (2023 (year only))
Z-FOLD: A Frustratingly Easy Post-Training Quantization Scheme for LLMs
2307.09288 (2023-07)
Llama 2: Open Foundation and Fine-Tuned Chat Models
and 24 more not named here
extract_alpaca_dataset 8 other papers
8 dated earlier
2305.14314 (2023-05)
QLORA: Efficient Finetuning of Quantized LLMs
2305.16300 (2023-05)
Random-Access Infinite Context Length for Transformers
2307.09288 (2023-07)
Llama 2: Open Foundation and Fine-Tuned Chat Models
and 5 more not named here
create_logger 7 other papers
7 dated earlier
2308.13137 (2023-08)
OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models
2401.09417 (2024-01)
Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Model
2403.12544 (2024-03)
AffineQuant: Affine Transformation Quantization for Large Language Models
and 4 more not named here

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

Mixed-precision quantization is a promising approach for compressing large language models under tight memory budgets. However, existing mixed-precision methods typically suffer from one of two limitations: they either rely on expensive discrete optimization to determine precision allocation, or introduce hardware inefficiencies due to irregular memory layouts. We propose SFMP, a search-free and hardwarefriendly mixed-precision quantization framework for large language models. SFMP defines a theoretically simple objective for mixedprecision quantization and is built upon three novel ideas: 1) Block-wise mixed-precision, enabling fine-grained precision within weight matrices while remaining hardware-friendly; 2) Row-column weight reordering, which improves structural alignment via row and column reordering, incurring only a small activation reordering overhead during inference; 3) Unified GEMM kernel, which supports blockwise mixed-precision GEMM at arbitrary average bit-widths. Extensive experiments demonstrate that SFMP outperforms state-of-the-art layer-wise mixed-precision methods under the same memory constraints, while significantly reducing quantization cost and improving inference efficiency. Code is available at https: //github.com/Nkniexin/SFMP.

For agents

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

get_harvested_code_for_paper("2602.01027")
get_code_for_paper("2602.01027")
have("2602.01027")

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

curl https://syntology.ai/api/ran/2602.01027.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.01027.svg)](https://syntology.ai/paper/2602.01027)

Connect an agent — have() is free.