datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Chameleon
Chameleon
Chameleon is a dataset designed for side-channel analysis of obfuscated power traces.
It contains real-world power traces collected from a 32-bit RISC-V System-on-Chip implementing four hiding countermeasures:
Dynamic Frequency Scaling (DFS), Random Delay (RD), Morphing (MRP), and Chaffing (CHF).
The dataset also includes side-channel power traces without any active countermeasure (BASE).
Each side-channel trace includes multiple cryptographic operations
interleaved… See the full description on the dataset page: https://huggingface.co/datasets/hardware-fab/Chameleon.unitree-g1-hardware-modifications
Unitree G1 Hardware Modifications
Custom 3D-printable parts and reference photos for modifying the Unitree G1: OpenArm
grippers on the wrists, an extra head camera, and a backpack enclosure for the CAN-FD interface
and cabling.
Everything here is printable geometry plus photos of the assembled result — there is no
firmware or control code in this repo.
Modifications
1. Gripper on the G1 wrist
The gripper itself is the stock gripper from the OpenArm… See the full description on the dataset page: https://huggingface.co/datasets/lerobot/unitree-g1-hardware-modifications.TroLL-Logic-Locking-based-Hardware-Trojanscross-hardware-reproducibility
Cross-hardware reproducibility of LLM evaluation results (item level)
Same model, same prompts, same decode settings, different machine: does each item get the same
answer? This dataset holds the item-level evidence behind the preprint "Same model, same prompts,
different answers: item-level cross-hardware reproducibility of LLM evaluation results"
(Nicholas Templeman, CSOAI Ltd (Council of AI), 2026). The PDF is in this repository:
paper/paper.pdf; its LaTeX source, with the… See the full description on the dataset page: https://huggingface.co/datasets/csoai/cross-hardware-reproducibility.openarms-hardware-modifications
OpenArm Hardware Modifications for Cloth Folding
Custom 3D-printable parts used in the Unfolding Robotics project, where we trained a bimanual robot to fold t-shirts with a 90% success rate.
These files modify the standard OpenArm.
Files
File
Description
J4_5cm_extended.step
Extended upper arm (bicep) segment, adds +5 cm of reach to compensate for the lack of a hip/torso in our setup. STEP format for easy modification.
J3-J4_Cover front extended.stl
Front… See the full description on the dataset page: https://huggingface.co/datasets/lerobot/openarms-hardware-modifications.DFS_DESYNCH
DFS_DESYNCH
The DFS_DESYNCH dataset contains power traces of a software AES implementation running on a 32-bit RISC-V System-on-Chip (SoC).
The SoC incorporates a Dynamic Frequency Scaling (DFS) unit that randomly adjusts the operating frequency between 35MHz and 60MHz.
Curated by: hardware-fab
License: Open Data Commons License cc-by-4.0
Repository: DLaTA methodology GitHub
Paper: A Deep Learning-assisted Template Attack Against Dynamic Frequency Scaling Countermeasures
The… See the full description on the dataset page: https://huggingface.co/datasets/hardware-fab/DFS_DESYNCH.hardware-cvdp-complete
CVDP - Comprehensive Verilog Design Problems (Complete Dataset)
🎯 782 out of 783 problems from the official CVDP benchmark by NVIDIA Research
🔥 Dataset Overview
This is the most complete version of the Comprehensive Verilog Design Problems (CVDP) benchmark available, containing 782 problems across 13 task categories. CVDP is designed to evaluate Large Language Models and agents on RTL design and verification tasks.
📊 Dataset Statistics
Total Problems: 772… See the full description on the dataset page: https://huggingface.co/datasets/AbiralArch/hardware-cvdp-complete.osiris
OSIRIS: Bridging Analog Layout Circuit Design and Machine Learning with Scalable Dataset Generation
OSIRIS is an end-to-end analog circuit design pipeline capable of producing, validating, and evaluating large volumes of layouts for generic analog circuits.
The OSIRIS 🤗 HuggingFace repository hosts the randomly generated dataset discussed in the paper.
This codebse provides the OSIRIS Python script (osiris.py) and the Python friendly version of the SkyWater 130nm PDK (SKY130_PDK)… See the full description on the dataset page: https://huggingface.co/datasets/hardware-fab/osiris.hardware_code_and_sec_smallhardware_code_and_sec_medianmodelfit-hardware-dataset
ModelFit: Local LLM Hardware Compatibility Dataset
An open dataset of which local AI models (Ollama) fit which hardware, by
parameter size, quantization, minimum RAM, and estimated memory load, across
Apple Silicon Macs, iPhones, and NVIDIA GPUs.
Maintained by ModelFit. Browse it as an interactive
table at modelfit.io/data; the canonical
machine-readable source is
modelfit.io/api/dataset.
156 models across 27 families (116 with a registry-verified local build, 40 cloud-only… See the full description on the dataset page: https://huggingface.co/datasets/modelfit/modelfit-hardware-dataset.diy-project-code-based-on-hardware-imageHoloMotion_hardwarehardware-cvdp-problems
Hardware Design AI Training Dataset
This dataset contains processed hardware design problems and Verilog code for training AI models.
Contents
CVDP Problems: 160 evaluation problems organized by domain and complexity
Training Data: Instruction-code pairs for hardware design
Metadata: Rich annotations for each problem
Usage
from datasets import load_dataset
dataset = load_dataset("AbiralArch/hardware-cvdp-problems")
Categories
Module Generation… See the full description on the dataset page: https://huggingface.co/datasets/AbiralArch/hardware-cvdp-problems.firewall-hardware-end-of-life-dates-by-brand
Firewall and network security appliance hardware end-of-life dates by brand
Canonical, always-current version: https://referencesource.org/firewall-hardware-end-of-life-dates-by-brand/
Machine-readable: https://referencesource.org/firewall-hardware-end-of-life-dates-by-brand/data.json — this mirror is a point-in-time copy.
Last verified: 2026-08-26
Stale after: 2027-02-22 (past this date, prefer the canonical copy —
it re-verifies on a cadence this snapshot does not)
Records:… See the full description on the dataset page: https://huggingface.co/datasets/referencesource/firewall-hardware-end-of-life-dates-by-brand.hardware-verilogeval-v2
hardware-verilogeval-v2
VerilogEval v2 - 471 Verilog evaluation problems
Dataset Overview
This dataset is part of a comprehensive collection of hardware design datasets for training and evaluating LLMs on Verilog/SystemVerilog code generation and hardware design tasks.
Files
verilog_eval_problems.json: 471 VerilogEval v2 problems
Usage
from datasets import load_dataset
# Load the dataset
dataset = load_dataset('AbiralArch/hardware-verilogeval-v2')… See the full description on the dataset page: https://huggingface.co/datasets/AbiralArch/hardware-verilogeval-v2.box_closing_v2.5_both_new_hardware_recomputed_statsThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"right_joint_1.pos",
"right_joint_2.pos",
"right_joint_3.pos",
"right_joint_4.pos",
"right_joint_5.pos",
"right_joint_6.pos",
"right_joint_7.pos"… See the full description on the dataset page: https://huggingface.co/datasets/Anteid11/box_closing_v2.5_both_new_hardware_recomputed_stats.r1-h4-trigger-hardware-15fpstheia
THEIA: A Multimodal Dataset and Benchmark for Vision-Language Analysis of Layout
NeurIPS 2026 Evaluations and Datasets
This repository contains the dataset presented in the paper "THEIA: A Multimodal Dataset and Benchmark for Vision-Language Analysis of Layout", along with the code for training and evaluating Visual Language Models (VLMs) on analog circuit layouts analysis tasks.
The project addresses the challenge of interpreting technical diagrams by benchmarking VLMs on tasks… See the full description on the dataset page: https://huggingface.co/datasets/hardware-fab/theia.africa-synth-telecom-hardware-sensor-data-nigeria
Africa Synth Telecom Hardware Sensor Data Nigeria | Africa (Electric Sheep Africa metadata inventory)
Size category: 100K<n<1M - Formats: parquet - Sector: energy - Engineered by Electric Sheep Africa
TL;DR
This dataset is part of the Electric Sheep Africa catalog on Hugging Face. It is indexed for African data discovery with standardized metadata, loading guidance, provenance notes, and analyst-oriented context.
What This Dataset Covers
Public… See the full description on the dataset page: https://huggingface.co/datasets/electricsheepafrica/africa-synth-telecom-hardware-sensor-data-nigeria.ai-inference-hardware-economics-2026
🚀 2026 AI Inference & Hardware Economics Telemetry Index
This repository hosts the official open-access empirical telemetry dataset for 2026 AI Inference, Silicon Architecture, and Hardware Economics, curated by EyesTech Systems & FinOps Intelligence.
Original Research Investigation:For the complete whitepaper, interactive latency calculators, and per-token TCO models, see the flagship publication at:👉… See the full description on the dataset page: https://huggingface.co/datasets/devidasmishra/ai-inference-hardware-economics-2026.digit3a_hardware_door_open_rgb_datasetThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "digit_third_arm",
"total_episodes": 33,
"total_frames": 13730,
"total_tasks": 1,
"total_videos": 33,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 50,
"splits": {
"train": "0:33"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/kelvinzhaozg/digit3a_hardware_door_open_rgb_dataset.Manufacturing-Hardware-and-MaterialsCommercial License Available: This is a free evaluation subset of a 71,777-image dataset. To acquire the full commercial dataset (100% clean-room guaranteed, zero PII, zero copyright risk), visit outpostvertical.com/manufacturing to purchase via Enterprise PO, or email data@outpostvertical.com for a SWIFT wire invoice.
EVALUATION SUBSET: The 2026 Manufacturing Hardware & Materials Vision Pack
1. Evaluation Dataset Overview
Eval Subset Image Count: 410 Images… See the full description on the dataset page: https://huggingface.co/datasets/outpostvertical/Manufacturing-Hardware-and-Materials.africa-synth-telecom-hardware-sensor-data-nigeria
Africa Synthetic Telecom Hardware Sensor Data Nigeria (TsFile)
This dataset is an Apache TsFile conversion of electricsheepafrica/africa-synth-telecom-hardware-sensor-data-nigeria, a synthetic Nigerian telecom tower hardware sensor dataset with temperature, power, voltage, humidity, vibration, health-status, and alert readings.
Source Dataset
Original dataset: electricsheepafrica/africa-synth-telecom-hardware-sensor-data-nigeria
Source files:… See the full description on the dataset page: https://huggingface.co/datasets/THULab/africa-synth-telecom-hardware-sensor-data-nigeria.modelfit-hardware-dataset
ModelFit Local LLM Hardware Compatibility Dataset
Which local AI models fit which hardware. Maps 107 LLMs (75 of them local-capable via Ollama, llama.cpp or LM Studio) to RAM/VRAM requirements at Q4_K_M quantization, so you can look up "will this model run on my machine" without guessing.
Source of truth: modelfit.io/data. This dataset is a mirror of the live JSON export at modelfit.io/api/dataset, refreshed from the same GitHub repo that generates it:… See the full description on the dataset page: https://huggingface.co/datasets/weckoai/modelfit-hardware-dataset.box_closing_v2.5_both_new_hardwareThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"names": [
"right_joint_1.pos",
"right_joint_2.pos",
"right_joint_3.pos",
"right_joint_4.pos",
"right_joint_5.pos",
"right_joint_6.pos",
"right_joint_7.pos"… See the full description on the dataset page: https://huggingface.co/datasets/Anteid11/box_closing_v2.5_both_new_hardware.quantum-hardware-device-physics
Neura Parse — Quantum Hardware Device Physics: Qubit Design, Coherence, Control & Scaling
A physics- and engineering-deep vertical on how qubits are built, controlled, and scaled across superconducting, trapped-ion, neutral-atom, and spin modalities (plus emerging erasure/biased-noise qubits). Device-physics derivations, coherence-limit analyses, control-stack engineering, and 2025-2026 scaling/interconnect work, with QuTiP/scqubits simulation context — expanding the general… See the full description on the dataset page: https://huggingface.co/datasets/Neura-parse/quantum-hardware-device-physics.ibm-150q-ising-hardware-results
IBM 150-Qubit Ising Hardware Results
Auditable results from a hardware-native, 150-active-qubit Ising/QAOA study on IBM ibm_fez, accompanied by an exact classical baseline and explicit scientific claim boundaries.
Headline evidence
Item
Verified value
Active physical qubits
150
Processor qubits
156
Successful hardware jobs
15
QPU usage
599 seconds
PUBs
547
Shots
2,087,936
QAOA depths
p=1 through p=8
Exact classical optimum
865 / 873, MIP… See the full description on the dataset page: https://huggingface.co/datasets/sankalpsthakur/ibm-150q-ising-hardware-results.ai-hardware-roadmap-2026
ai-hardware-roadmap-2026
AI chip roadmap — H100/H200/B200/MI300X/GB200 specs, pricing, availability
Records: 15 | Updated: 2026-10-01
API: GET https://api.legion-api.com/hardware-roadmap
Bundle: gemmo.gumroad.com/l/mdevxu
Access gated — approved automatically.
digit3a_shelf_hardwareThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "digit_third_arm",
"total_episodes": 27,
"total_frames": 35085,
"total_tasks": 1,
"total_videos": 0,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 12.5,
"splits": {
"train": "0:27"
},
"data_path": "data/chunk-{episode_chunk:03d}/episode_{episode_index:06d}.parquet",
"video_path": null,
"features": {… See the full description on the dataset page: https://huggingface.co/datasets/kelvinzhaozg/digit3a_shelf_hardware.