datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
NECL_GPUskernelbot-data
KernelBot Competition Data
This dataset contains GPU kernel submissions from the KernelBot competition platform. Submissions are optimized GPU kernels written for specific hardware targets.
Data Files
AMD MI300 Submissions
File
Description
submissions.parquet
All AMD competition submissions
successful_submissions.parquet
AMD submissions that passed correctness tests
deduplicated_submissions.parquet
AMD submissions deduplicated by… See the full description on the dataset page: https://huggingface.co/datasets/GPUMODE/kernelbot-data.backendbench_tests
TorchBench
The TorchBench suite of BackendBench is designed to mimic real-world use cases. It provides operators and inputs derived from 155 model traces found in TIMM (67), Hugging Face Transformers (45), and TorchBench (43). (These are also the models PyTorch developers use to validate performance.) You can view the origin of these traces by switching the subset in the dataset viewer to ops_traces_models and torchbench for the full dataset.
When running BackendBench, much of the… See the full description on the dataset page: https://huggingface.co/datasets/GPUMODE/backendbench_tests.gpu-database
GPU Database
Comprehensive GPU specifications database with architecture, manufacturing, API support, performance details, and kernel development specs.
2,824 GPUs across NVIDIA, AMD, and Intel
Part of RightNow — AI-powered code editor for GPU kernel development
Data
Vendor
GPUs
File
NVIDIA
1,286
data/nvidia/all.json
AMD
1,292
data/amd/all.json
Intel
180
data/intel/all.json
All
2,824
data/all-gpus.json
Schema
Each GPU contains up to 55… See the full description on the dataset page: https://huggingface.co/datasets/Jr23xd23/gpu-database.KernelBook
Overview
dataset_permissive{.json/.parquet} is a curated collection of pairs of pytorch programs and equivalent triton code (generated by torch inductor) which can be used to train models to translate pytorch code to triton code.
The triton code was generated using PyTorch 2.5.0 so for best results during evaluation / running the triton code we recommend using that version of pytorch.
Dataset Creation
The dataset was created through the following process:… See the full description on the dataset page: https://huggingface.co/datasets/GPUMODE/KernelBook.so101_pick_cube
SO-101 Pick Cube Dataset
A large-scale robotics manipulation dataset featuring the SO-101 robot arm performing pick-and-place tasks in MuJoCo simulation. The robot learns to pick up a cube and place it into a bin.
Dataset Details
Property
Value
Robot
SO-101 (6-DoF arm with gripper)
Task
Pick cube and place in bin
Episodes
10,993
Total Frames
1,456,901
FPS
30
Cameras
3 (front, overhead, wrist)
Resolution
640x480
Format
LeRobot v3.0… See the full description on the dataset page: https://huggingface.co/datasets/gpudad/so101_pick_cube.cloud-gpu-price-index
Cloud GPU Price Index
Canonical source: https://gpueconomy.com/price-index. That page is
recomputed every hour; this record is a dated snapshot of it, version
2026-10-02, built from data updated 2026-10-02T13:39:58.768836+00:00. When you cite,
cite GPU Economy and link the page; the snapshot is here so that a number you
used keeps existing exactly as you used it.
The index is the weekly median publicly listed on-demand price of one NVIDIA
H100 SXM GPU-hour across the cloud GPU… See the full description on the dataset page: https://huggingface.co/datasets/gpueconomy/cloud-gpu-price-index.eval_omx_act_gpu_30kThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "omx_follower",
"total_episodes": 7,
"total_frames": 3040,
"total_tasks": 1,
"total_videos": 14,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:7"
},
"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/jetsonmom/eval_omx_act_gpu_30k.one-layer-deeper-submissions
One Layer Deeper submissions
This dataset archives 15,602 accepted uploads from 206 GitHub accounts to the One Layer Deeper competition. It contains 9,627 distinct source files, all upload metadata, and stored evaluation results. Snapshot: September 7, 2026, 21:48 UTC, after the August 31 submission deadline.
Split
Uploads
Succeeded
Failed
easy
11,961
11,112
849
medium
2,704
2,509
195
hard
937
847
90
All accepted uploads are included: practice runs, failures… See the full description on the dataset page: https://huggingface.co/datasets/GPUMODE/one-layer-deeper-submissions.so101_pick_cube_chunked
SO101 Pick Cube Dataset (Chunked)
This is a restructured version of the gpudad/so101_pick_cube dataset with episode-level video files for faster data loading during training.
Why Chunked?
The original dataset has 3 monolithic video files (one per camera, 13+ hours each). Random access during training is slow because the decoder must seek through huge files.
This version splits videos into 1000 episodes per chunk, making data loading ~50x faster.
Dataset Info… See the full description on the dataset page: https://huggingface.co/datasets/gpudad/so101_pick_cube_chunked.pc_gpu_franka_ik.jointtarget_20hz_20260906This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 20,
"features": {
"observation.images.front": {
"dtype": "video",
"shape": [
480,
640,
3
],
"names": [
"height",
"width",
"channels"
],
"info": {
"video.height":… See the full description on the dataset page: https://huggingface.co/datasets/EmbodiedSWE/pc_gpu_franka_ik.jointtarget_20hz_20260906.so101_gpulearntestThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "so101",
"total_episodes": 20,
"total_frames": 5976,
"total_tasks": 1,
"total_videos": 40,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:20"
},
"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/ganondorofu/so101_gpulearntest.l4-gpu-llm-benchmark-leaderboard
🚀 Local LLM Serving & Quality Benchmark Leaderboard (NVIDIA L4 24GB)
An exhaustive, reproducible benchmark study measuring real-world serving performance (TTFT, TPOT, throughput, peak VRAM, energy consumption, and cost) alongside rigorous task quality gates (HumanEval+, MMLU-Pro, BFCL v4 tool calling, and RULER needle retrieval) for open-weight LLMs on a single NVIDIA L4 24GB GPU.
📊 Executive Summary & Key Takeaways
⚡ Best Throughput & Coding Workhorse:… See the full description on the dataset page: https://huggingface.co/datasets/mayank-dubey-ai/l4-gpu-llm-benchmark-leaderboard.gpuark-gpu-dataset
GPU Ark — open GPU specifications & benchmarks dataset
Specifications of 13,566 GPUs released between 1999 and 2025 — from the GeForce 256 to
NVIDIA Blackwell and AMD Instinct MI355X — plus 993 third-party benchmark results.
Curated and maintained by GPU Ark (a GPU catalog & price comparison
project). Canonical source and always-fresh copy: https://gpuark.com/datasets/.
Files
File
Rows
What
gpuark-gpu-specs.csv
13,566
One row per GPU — public spec columns… See the full description on the dataset page: https://huggingface.co/datasets/Intelion/gpuark-gpu-dataset.gpu-compatibility
Self-Hosted AI — GPU Compatibility, Recipes and Catalogue
Which open-weight AI models actually run on which consumer GPU, and what it takes
to get them running. 2 700 model×GPU verdicts across 100 models
and 27 cards, plus 1 009 full setup guides
(19 MB of markdown) written against specific hardware.
This is the machine-readable form of smeltcore.com. Every row carries a
url back to the page it came from.
Generated 2026-09-24T19:45:21+00:00 from the public read API… See the full description on the dataset page: https://huggingface.co/datasets/Smeltcore/gpu-compatibility.pc_gpu_ram_franka_osc.jointpos_15hz_20260907This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "assembly.pc_gpu_ram.franka.osc",
"total_episodes": 600,
"total_frames": 1400130,
"total_tasks": 3,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 78,
"fps": 15,
"splits": {
"train": "0:600"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet"… See the full description on the dataset page: https://huggingface.co/datasets/EmbodiedSWE/pc_gpu_ram_franka_osc.jointpos_15hz_20260907.gpu-forecasters-puct-search-eventsCompanion artifact for GPU Forecasters: Language Models as Selective Surrogates for Kernel Runtime Optimization. Code: codezakh/gpu-surrogates.
Events emitted during the kernel searches in this work. You can reconstruct everything that happened in a search from these events: which kernels were tried, in what order, with what runtimes.
See the kernel-search code at codezakh/gpu-surrogates for the payload schema of each event kind.
Loading
from datasets import load_dataset
#… See the full description on the dataset page: https://huggingface.co/datasets/codezakh/gpu-forecasters-puct-search-events.Bunny_pic_up_20261007_113307This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"fps": 30,
"features": {
"action": {
"dtype": "float32",
"shape": [
6
],
"names": [
"shoulder_pan.pos",
"shoulder_lift.pos",
"elbow_flex.pos",
"wrist_flex.pos",
"wrist_roll.pos",
"gripper.pos"… See the full description on the dataset page: https://huggingface.co/datasets/GPUIQ/Bunny_pic_up_20261007_113307.nova_ur5_demoThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "nova_sim",
"total_episodes": 92,
"total_frames": 30325,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 10,
"splits": {
"train": "0:92"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gpue/nova_ur5_demo.gpu-forecasters-eval-setCompanion artifact for GPU Forecasters: Language Models as Selective Surrogates for Kernel Runtime Optimization. Code: codezakh/gpu-surrogates.
Held-out evaluation set used in the paper. Each row is one (reference, candidate) kernel pair on a GPU Mode task, with the candidate's measured speedup over the reference on an A100.
Loading
from datasets import load_dataset
# all six packs combined
ds = load_dataset("codezakh/gpu-forecasters-eval-set", name="combined", split="eval")
#… See the full description on the dataset page: https://huggingface.co/datasets/codezakh/gpu-forecasters-eval-set.pc_gpu_franka_osc.jointpos_15hz_20260907This dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "assembly.pc_gpu.franka.osc",
"total_episodes": 600,
"total_frames": 1128960,
"total_tasks": 5,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 81,
"fps": 15,
"splits": {
"train": "0:600"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet"… See the full description on the dataset page: https://huggingface.co/datasets/EmbodiedSWE/pc_gpu_franka_osc.jointpos_15hz_20260907.autonomous-cloud-gpu-slurm-serving-suite
⚡ Autonomous Cloud GPU Infrastructure, Slurm Orchestration & Distributed Serving Suite (2026)
A Production-Grade, Verifiable Synthetic Corpus for Training Autonomous AI Supercomputing & LLM Serving Agents
⚡ Overview & Industry Problem
Operating massive AI supercomputers (thousands of NVIDIA H100/H200 and Blackwell GPUs) requires coordinating Slurm cluster schedules, topology-aware NVLink cliques, NCCL AllReduce rings, RoCE v2 lossless fabrics… See the full description on the dataset page: https://huggingface.co/datasets/beatsprom/autonomous-cloud-gpu-slurm-serving-suite.autonomous-gpu-kernel-triton-cuda-suite-2026
⚡ Autonomous GPU Kernel, Triton & CUDA Architecture Suite (2026)
A Production-Grade, Verifiable Synthetic Corpus for Training Frontier Coding Models (Qwen 3.8, DeepSeek-V3, Llama 3.3)
⚡ Overview & Industry Problem
Modern deep learning accelerators, custom ASICs, and high-performance computing clusters demand specialized, autonomous GPU kernel infrastructure: OpenAI Triton fused kernels, FlashAttention-3 forward/backward online softmax… See the full description on the dataset page: https://huggingface.co/datasets/beatsprom/autonomous-gpu-kernel-triton-cuda-suite-2026.cuda-triton-gpu-kernels-2026
⚡ Complete 2026 CUDA & OpenAI Triton High-Performance GPU Kernel Engineering SFT/DPO Suite
The definitive, production-grade synthetic alignment dataset engineered for training and fine-tuning open-weights Large Language Models (Qwen 2.5 Coder, DeepSeek-Coder, Llama 3.1) on ultra-high-throughput GPU kernel programming: NVIDIA Hopper H100 / Blackwell B200 TMA async transfers, OpenAI Triton 3.1+ FlashAttention-3, 32-bank conflict elimination, and low-bit FP8 / INT4 GEMM… See the full description on the dataset page: https://huggingface.co/datasets/beatsprom/cuda-triton-gpu-kernels-2026.so101_pick_cube_v2
SO-101 Pick Cube Dataset
A robotics manipulation dataset featuring the SO-101 robot arm performing pick-and-place tasks in MuJoCo simulation. The robot learns to pick up a cube and place it into a bin.
Dataset Details
Property
Value
Robot
SO-101 (6-DoF arm with gripper)
Task
Pick cube and place in bin
Episodes
993
FPS
30
Cameras
3 (front, overhead, wrist)
Resolution
640x480
Format
LeRobot v2.0
Cameras
Front Camera:… See the full description on the dataset page: https://huggingface.co/datasets/gpudad/so101_pick_cube_v2.cloud-gpu-prices
Cloud GPU Rental Prices (FastGPU)
What it costs to rent a GPU in the cloud, per GPU per hour, across marketplaces, neoclouds and hyperscalers. Mirrored here once a day from fastgpu.co/dataset, where the same files are served live and every price links to the provider it came from. Browse the prices themselves at fastgpu.co.
A free, openly licensed dataset of GPU cloud rental prices across the whole market: marketplaces, neoclouds, and hyperscalers. It contains a normalized… See the full description on the dataset page: https://huggingface.co/datasets/fastgpu/cloud-gpu-prices.pallasbench-robust-gpu-a100
PallasBench: Robust Pallas GPU Kernel Benchmark (A100)
39/45 kernels passing on NVIDIA A100 80GB -- the first GPU-focused evaluation of JAX Pallas kernels.
What is this?
PallasBench is a suite of 45 JAX Pallas kernels across 3 difficulty levels. The original kernels were designed for TPU and failed on GPU because Pallas compiles to Triton on NVIDIA hardware, which has strict block size limits that TPU's Mosaic compiler does not.
We fixed all 45 kernels for GPU… See the full description on the dataset page: https://huggingface.co/datasets/EvanOLeary/pallasbench-robust-gpu-a100.kanvas-gpu-failure
KANVAS GPU Task Failure Features
Derived data for the KANVAS project: an interpretable Kolmogorov-Arnold Additive Model (KAAM) that predicts GPU cluster task failure. Code, notebooks and tests: https://github.com/Swapnil007-Curious/KANVAS
Files
File
Rows
Contents
kanvas_feature_matrix.csv
18,890
10 scheduling and telemetry features plus the failure label, one row per task
kanvas_phi_curves.csv
2,000
Learned per-feature curves phi_i(x) of the canonical… See the full description on the dataset page: https://huggingface.co/datasets/Swapnil007-Curious/kanvas-gpu-failure.eval_act_so100_obj_to_bin_top0_180_gpuThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v2.1",
"robot_type": "so100",
"total_episodes": 5,
"total_frames": 3014,
"total_tasks": 1,
"total_videos": 10,
"total_chunks": 1,
"chunks_size": 1000,
"fps": 30,
"splits": {
"train": "0:5"
},
"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/taxitain/eval_act_so100_obj_to_bin_top0_180_gpu.ur5-followThis dataset was created using LeRobot.
Dataset Structure
meta/info.json:
{
"codebase_version": "v3.0",
"robot_type": "nova_sim",
"total_episodes": 100,
"total_frames": 6145,
"total_tasks": 1,
"chunks_size": 1000,
"data_files_size_in_mb": 100,
"video_files_size_in_mb": 200,
"fps": 20,
"splits": {
"train": "0:100"
},
"data_path": "data/chunk-{chunk_index:03d}/file-{file_index:03d}.parquet",
"video_path":… See the full description on the dataset page: https://huggingface.co/datasets/gpue/ur5-follow.