SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.07597 · 2021

MLPerf Tiny Benchmark

arXiv · PDF · Open in the Atlas

Code that ran

9 of 16 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 8 have no contract verdict; 0 failures of our instrument; 7 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/2106.07597.json · badge: /api/ran/2106.07597.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
mlcommons/tiny canonical (classifier label)
classifier basis: census
7 ran · 4 no recorded run
mlcommons/tiny_results_v0.5 canonical (classifier label)
classifier basis: census
2 ran · 3 no recorded run
FunctionStatusWhere it lives
get_baud_rate Ran
ran on a synthesized input
mlcommons/tiny/benchmark/runner/runner_utils.py
code served (permissive licence) · get_code("24bf521db881643a")
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 int, ndim 0, dtype family int, 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 torch
import numpy as np

# Construct a valid dut_config with baud rate information
dut_config_simple = {
    'baud': 9600,
    'name': 'device_1'
}

dut_config_mode_based = {
    'baud': {
        'energy': 4800,
        'performance': 115200,
        'accuracy': 9600
    },
    'name': 'device_2'
}

# Test with simple baud rate (integer)
ARGS = {
    'dut_config': dut_config_simple,
    'mode': None
}
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 int, ndim 0, dtype family int, 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_model Ran
ran on a synthesized input
mlcommons/tiny/benchmark/training/anomaly_detection/keras_model.py
code served (permissive licence) · get_code("69600599471ffc0c")
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 Functional. 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 tensorflow as tf
from tensorflow.keras.models import Model
from tensorflow.keras.layers import Input, Dense, BatchNormalization, Activation

# Set seed for reproducibility
tf.random.set_seed(0)

# The function takes a single parameter: inputDim
# This is an integer representing the input dimension
ARGS = {
    "inputDim": 4
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional 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
get_name Ran
ran on a synthesized input
mlcommons/tiny/benchmark/runner/stream_wav_uart.py
code served (permissive licence) · get_code("52eb372fa50a6b62")
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 str. 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 time
import io

class MockSerial:
    """Mock serial port that simulates device responses"""
    def __init__(self):
        self.response_data = "device-name-m-ready"
        self.write_buffer = []
        self.read_pos = 0
    
    def write(self, data):
        """Mock write - stores what was written"""
        self.write_buffer.append(data)
        return len(data)
    
    def read(self, num_bytes):
        """Mock read - returns response data one byte at a time"""
        if self.read_pos < len(self.response_data):
            char = self.response_data[self.read_pos].encode()
            self.read_pos += 1
            return char
        return b''

port = MockSerial()

ARGS = {
    'port': port
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
load_model Ran
ran on a synthesized input
mlcommons/tiny/benchmark/training/anomaly_detection/keras_model.py
code served (permissive licence) · get_code("62ef34c2f0401b44")
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 Sequential. 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 tensorflow.keras as keras
import tempfile
import os

# Create a minimal valid Keras model and save it to a temporary file
def create_and_save_minimal_model():
    # Build a simple sequential model
    model = keras.Sequential([
        keras.layers.Dense(4, activation='relu', input_shape=(3,)),
        keras.layers.Dense(2, activation='softmax')
    ])
    model.compile(optimizer='adam', loss='sparse_categorical_crossentropy')
    
    # Save to temporary file
    temp_dir = tempfile.mkdtemp()
    file_path = os.path.join(temp_dir, 'model.keras')
    model.save(file_path)
    
    return file_path

file_path = create_and_save_minimal_model()

ARGS = {
    'file_path': file_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential 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_model_settings Ran
ran on a synthesized input
mlcommons/tiny_results_v0.5/closed/PengChengLaboratory/code/ad/tiny/v0.1/training/keyword_spotting/keras_model.py
code served (permissive licence) · get_code("b35b49d0dc790d3d")
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 dict. 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 numpy as np

torch.manual_seed(0)

ARGS = {
    'label_count': 10,
    'sample_rate': 16000,
    'clip_duration_ms': 1000,
    'window_size_ms': 30,
    'window_stride_ms': 10,
    'dct_coefficient_count': 40,
    'background_frequency': 0.8
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
process_dutycycle Ran
ran on a synthesized input
mlcommons/tiny/benchmark/runner/streaming_ww_utils.py
code served (permissive licence) · get_code("80bcdb8ca9416120")
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 dict. 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 numpy as np

ARGS = {
    'raw_result': [
        'Duty cycle start times (s)',
        '0,100,200,300',
        '400,500,600',
        'Duty cycle stop times (s)',
        '50,150,250,350',
        '450,550,650,m-ready'
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
process_timestamps Ran
ran on a synthesized input
mlcommons/tiny/benchmark/runner/streaming_ww_utils.py
code served (permissive licence) · get_code("53c9956822b2012c")
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 list, ndim 1, dtype family int; 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

# Construct valid arguments for process_timestamps
raw_result = [
    'Detection Timestamps (ms)',
    '100,',
    '250,',
    '500,',
    '1000,',
    'm-ready'
]

ARGS = {
    'raw_result': raw_result
}
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 list, ndim 1, dtype family int — a list 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
resnet_v1_eembc Ran
ran on a synthesized input
mlcommons/tiny/benchmark/training/image_classification/keras_model.py
code served (permissive licence) · get_code("958a1150a034c5e4")
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 Functional. 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 numpy as np
import tensorflow as tf

ARGS = {
    "conv_filters": 26
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional 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
str2bool Ran
ran, violated the contract we drafted
mlcommons/tiny_results_v0.5/closed/PengChengLaboratory/code/ad/ConvertModel.py
code served (permissive licence) · get_code("7c508037b40522af")
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 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 argparse

def str2bool(v):
    if isinstance(v, bool):
        return v
    if v.lower() in ('yes', 'true', 't', 'y', '1'):
        return True
    elif v.lower() in ('no', 'false', 'f', 'n', '0'):
        return False
    else:
        raise argparse.ArgumentTypeError('Boolean value expected.')

# Test cases for str2bool
ARGS = {
    'v': 'true'  # string input that should convert to boolean
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
array_from_strings Not yet run
unverified (harvested, no recorded run)
mlcommons/tiny/benchmark/runner/streaming_ww_utils.py
code served (permissive licence) · get_code("7340ad06fb362e40")
How it ran

no invocation recorded.

effnet_v2s Not yet run
unverified (harvested, no recorded run)
mlcommons/tiny/benchmark/training/image_classification/keras_model.py
code served (permissive licence) · get_code("85ee9507d23e68c0")
How it ran

no invocation recorded.

get_model Not yet run
unverified (harvested, no recorded run)
mlcommons/tiny_results_v0.5/closed/PengChengLaboratory/code/ad/tiny/v0.1/training/keyword_spotting/keras_model.py
code served (permissive licence) · get_code("c3a0293fcad9180a")
How it ran

no invocation recorded.

get_response Not yet run
unverified (harvested, no recorded run)
mlcommons/tiny/benchmark/runner/stream_wav_uart.py
code served (permissive licence) · get_code("ffecdb9ac0a5c52f")
How it ran

no invocation recorded.

predict Not yet run
unverified (harvested, no recorded run)
mlcommons/tiny_results_v0.5/closed/PengChengLaboratory/code/ad/tiny/v0.1/training/keyword_spotting/eval_quantized_model.py
code served (permissive licence) · get_code("fb79e3cbc6774c55")
How it ran

no invocation recorded.

send_cmd_get_resp Not yet run
unverified (harvested, no recorded run)
mlcommons/tiny/benchmark/runner/stream_wav_uart.py
code served (permissive licence) · get_code("b5b3bd1958d30306")
How it ran

no invocation recorded.

toByteBuffer Not yet run
unverified (harvested, no recorded run)
mlcommons/tiny_results_v0.5/closed/PengChengLaboratory/code/ad/ConvertModel.py
code served (permissive licence) · get_code("54aa90f290420a50")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 1 later) and 15 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
str2bool 175 other papers
32 dated earlier, 4 the same month, 139 later · 2 only as an assembly brick
1406.6247 (2014-06)
Recurrent Models of Visual Attention
1603.09382 (2016-03)
Deep Networks with Stochastic Depth
1710.10196 (2017-10)
Progressive Growing of GANs for Improved Quality, Stability, and Variation
and 172 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

Advancements in ultra-low-power tiny machine learning (TinyML) systems promise to unlock an entirely new class of smart applications. However, continued progress is limited by the lack of a widely accepted and easily reproducible benchmark for these systems. To meet this need, we present MLPerf Tiny, the first industry-standard benchmark suite for ultra-low-power tiny machine learning systems. The benchmark suite is the collaborative effort of more than 50 organizations from industry and academia and reflects the needs of the community. MLPerf Tiny measures the accuracy, latency, and energy of machine learning inference to properly evaluate the tradeoffs between systems. Additionally, MLPerf Tiny implements a modular design that enables benchmark submitters to show the benefits of their product, regardless of where it falls on the ML deployment stack, in a fair and reproducible manner. The suite features four benchmarks: keyword spotting, visual wake words, image classification, and anomaly detection.

For agents

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

get_harvested_code_for_paper("2106.07597")
get_code_for_paper("2106.07597")
have("2106.07597")

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

curl https://syntology.ai/api/ran/2106.07597.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.07597.svg)](https://syntology.ai/paper/2106.07597)

Connect an agent — have() is free.