SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2411.05282 · 2024

MicroScopiQ: Accelerating Foundational Models through Outlier-Aware Microscaling Quantization

arXiv · PDF · Open in the Atlas

Code that ran

6 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, 1 violated it and 0 have no contract verdict; 5 failures of our instrument; 6 no recorded run), measured 2026-11-21; 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/2411.05282.json · badge: /api/ran/2411.05282.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
georgia-tech-synergy-lab/microscopiq-llm-quantization canonical (classifier label)
classifier basis: census
6 ran · 6 no recorded run
FunctionStatusWhere it lives
analyze_outliers_olive Ran
ran; our contract draft was wrong, not the code
georgia-tech-synergy-lab/microscopiq-llm-quantization/outlier_profiling/profiler.py
code served (permissive licence) · get_code("b98e145c4c83940d")
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 torch
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
        self.proj = nn.Linear(2, 2)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        x = self.proj(x)
        return x

model = SimpleModel()

ARGS = {
    "model": model
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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
compare_models_and_check_nans Ran
ran; our contract draft was wrong, not the code
georgia-tech-synergy-lab/microscopiq-llm-quantization/llm/opt_direct.py
code served (permissive licence) · get_code("88176b06dfb6df3e")
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, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self, input_dim=4, hidden_dim=3, output_dim=2):
        super().__init__()
        self.linear1 = nn.Linear(input_dim, hidden_dim)
        self.linear2 = nn.Linear(hidden_dim, output_dim)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

# Create two identical models
model1 = SimpleModel(input_dim=4, hidden_dim=3, output_dim=2)
model2 = SimpleModel(input_dim=4, hidden_dim=3, output_dim=2)

# Copy parameters from model1 to model2 so they are identical
with torch.no_grad():
    for param1, param2 in zip(model1.parameters(), model2.parameters()):
        param2.copy_(param1)

ARGS = {
    'model1': model1,
    'model2': model2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
evaluate_pred_answer Ran
ran; our contract draft was wrong, not the code
georgia-tech-synergy-lab/microscopiq-llm-quantization/kv_quant/evaluation_gsm8k.py
code served (permissive licence) · get_code("b6a8148f2f12adc9")
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 re

def evaluate_pred_answer(pred_str, ans_str):
    pattern = '\\d*\\.?\\d+'
    pred_str, ans_str = (pred_str.replace(',', ''), ans_str.replace(',', ''))
    pred_list = re.findall(pattern, pred_str)
    gold_list = re.findall(pattern, ans_str)
    if len(pred_list) >= 1:
        pred = float(pred_list[-1])
        gold = float(gold_list[-1])
        is_pred_true = pred == gold
    else:
        is_pred_true = False
        pred = None
        gold = float(gold_list[-1])
    return (is_pred_true, pred, pred_list, gold, gold_list)

ARGS = {
    'pred_str': 'The answer is 42.5',
    'ans_str': 'The correct answer is 42.5'
}
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 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
find_layers Ran
ran; our contract draft was wrong, not the code
georgia-tech-synergy-lab/microscopiq-llm-quantization/utils/modelutils.py
pointer only (licence: MIT); 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: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from squeezeailab/kvquant, and GitHub detected no licence in that repository 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_opt Ran
ran; our contract draft was wrong, not the code
georgia-tech-synergy-lab/microscopiq-llm-quantization/llm/opt.py
code served (permissive licence) · get_code("887f43fe08a6472d")
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) and returned MockOPTForCausalLM. 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 torch.nn as nn

torch.manual_seed(0)

# The function expects a model parameter that is a string (model name/path)
# However, since we can't actually download from HuggingFace, we need to mock
# the from_pretrained call. We'll create a minimal mock setup.

class MockConfig:
    def __init__(self):
        self.max_position_embeddings = 2048

class MockOPTForCausalLM(nn.Module):
    def __init__(self):
        super().__init__()
        self.config = MockConfig()
        self.linear = nn.Linear(4, 4)
    
    @classmethod
    def from_pretrained(cls, model_name, torch_dtype='auto'):
        return cls()

# Monkey patch the transformers module before the function runs
import sys
from types import ModuleType

mock_transformers = ModuleType('transformers')
mock_transformers.OPTForCausalLM = MockOPTForCausalLM
sys.modules['transformers'] = mock_transformers

# The model parameter should be a string (model identifier)
model_name = "facebook/opt-125m"

ARGS = {
    "model": model_name
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type MockOPTForCausalLM — a MockOPTForCausalLM output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockOPTForCausalLM 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
test_answer Ran
ran, violated the contract we drafted
georgia-tech-synergy-lab/microscopiq-llm-quantization/kv_quant/evaluation_gsm8k.py
code served (permissive licence) · get_code("b6065f749757a201")
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 bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 re

def test_answer(pred_str, ans_str):
    pattern = '\\d*\\.?\\d+'
    pred = re.findall(pattern, pred_str)
    if len(pred) >= 1:
        print('#####\n Pred string:', pred_str, '\n pred_list', pred)
        pred = float(pred[-1].replace(',', ''))
        gold = re.findall(pattern, ans_str)
        print('\n Gold_answer', ans_str, '\n gold_list', gold)
        gold = float(gold[-1].replace(',', ''))
        print('\n result', gold, pred, gold == pred)
        return pred == gold
    else:
        return False


ARGS = {
    'pred_str': 'The answer is 42.5',
    'ans_str': 'The correct answer is 42.5'
}
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 bool, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
get_llama Not yet run
unverified (harvested, no recorded run)
georgia-tech-synergy-lab/microscopiq-llm-quantization/llm/llama.py
code served (permissive licence) · get_code("d1d25542bf6a73b0")
How it ran

no invocation recorded.

layer_norm Not yet run
unverified (harvested, no recorded run)
georgia-tech-synergy-lab/microscopiq-llm-quantization/number_system/mx/layernorm.py
code served (permissive licence) · get_code("494a1f0c452b05c6")
How it ran

no invocation recorded.

llama_sequential Not yet run
unverified (harvested, no recorded run)
georgia-tech-synergy-lab/microscopiq-llm-quantization/llm/llama.py
code served (permissive licence) · get_code("c6cb1685a5860766")
How it ran

no invocation recorded.

opt_pack3 Not yet run
unverified (harvested, no recorded run)
georgia-tech-synergy-lab/microscopiq-llm-quantization/llm/opt.py
code served (permissive licence) · get_code("5a1ab8a11275fe3d")
How it ran

no invocation recorded.

opt_sequential Not yet run
unverified (harvested, no recorded run)
georgia-tech-synergy-lab/microscopiq-llm-quantization/llm/opt.py
code served (permissive licence) · get_code("3ef2f25d56625af9")
How it ran

no invocation recorded.

parse_pred_ans Not yet run
unverified (harvested, no recorded run)
georgia-tech-synergy-lab/microscopiq-llm-quantization/kv_quant/evaluation_gsm8k.py
code served (permissive licence) · get_code("d93eb10d932ff2c8")
How it ran

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, 7 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 1 the same month, 2 later) and 5 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
20 dated earlier, 1 the same month, 4 later, 1 same year, order not determinable, 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
get_opt 5 other papers
4 dated earlier, 1 later
2210.17323 (2022-10)
GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
2301.00774 (2023-01)
SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot
2307.13304 (2023-07)
QuIP: 2-Bit Quantization of Large Language Models With Guarantees
and 2 more not named here
opt_pack3 2 other papers
2 dated earlier
2210.17323 (2022-10)
GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers
2406.05981 (2024-06)
ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization
evaluate_pred_answer 1 other paper
1 dated earlier
2403.05527 (2024-03)
GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM
layer_norm 1 other paper
1 dated earlier
2310.10537 (2023-10)
Microscaling Data Formats for Deep Learning
parse_pred_ans 1 other paper
1 dated earlier
2403.05527 (2024-03)
GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM
test_answer 1 other paper
1 dated earlier
2403.05527 (2024-03)
GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Quantization of foundational models (FMs) is significantly more challenging than traditional DNNs due to the emergence of large magnitude values called outliers. Existing outlier-aware algorithm-architecture co-design techniques either use mixed-precision, retaining outliers at high precision but compromise hardware efficiency, or quantize inliers and outliers at the same precision, improving hardware efficiency at the cost of accuracy. To address this mutual exclusivity, we propose MicroScopiQ, a novel co-design technique that leverages pruning to complement outlier-aware quantization. MicroScopiQ retains outliers at higher precision while pruning a certain fraction of least important weights to distribute the additional outlier bits; ensuring high accuracy, aligned memory and hardware efficiency. We design a high-throughput, low overhead accelerator architecture composed of multi-precision INT processing elements and a network-on-chip called ReCoN that efficiently abstracts the complexity of supporting high-precision outliers. Additionally, unlike prior techniques, MicroScopiQ does not assume any locality of outlier weights, enabling applicability to a broad range of FMs. Extensive experiments across diverse quantization settings demonstrate that MicroScopiQ achieves state-of-the-art quantization accuracy, while delivering up to 3x faster inference and 2x lower energy consumption compared to existing alternatives. Code is available at: https://github.com/georgia-tech-synergy-lab/MicroScopiQ-LLM-Quantization

For agents

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

get_harvested_code_for_paper("2411.05282")
get_code_for_paper("2411.05282")
have("2411.05282")

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

curl https://syntology.ai/api/ran/2411.05282.json

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

[![Syntology run record](https://syntology.ai/api/ran/2411.05282.svg)](https://syntology.ai/paper/2411.05282)

Connect an agent — have() is free.