datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
CUDA-Agent-Ops-6K
CUDA-Agent-Ops-6K
CUDA-Agent-Ops-6K is a curated training dataset for CUDA kernel generation and optimization.
It is released as part of the CUDA-Agent project:
Project Page: https://CUDA-Agent.github.io/
Github Repo: https://github.com/BytedTsinghua-SIA/CUDA-Agent
Dataset Summary
CUDA-Agent-Ops-6K contains 6,000 synthesized operator-level training tasks designed for large-scale agentic RL training. It is intended to provide diverse and executable CUDA-oriented training… See the full description on the dataset page: https://huggingface.co/datasets/BytedTsinghua-SIA/CUDA-Agent-Ops-6K.Nemotron-SFT-CUDA-v1
Dataset Description:
Nemotron-SFT-CUDA-v1 is a training dataset for CUDA code. It helps language models write CUDA kernels and solve CUDA programming problems.
We start from CUDA code in Nemotron Pretraining Code v2, which has a permissive license. An OpenCode agent powered by GLM-4.7 reads that code and writes new CUDA programming problems. Each problem comes with a hidden answer and tests. A second OpenCode + GLM-4.7 agent then tries to solve the problems using only the prompt… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-SFT-CUDA-v1.r8-eval-suite-5bucket
⚠️ CRITICAL: Ollama Inference Flag Required for derived models
If you train or serve any Qwen3.5-9B-derived model from this lineage via Ollama,
you MUST pass "think": false in /api/chat requests for chat / instruction following / tool use.
The qwen3.5 RENDERER auto-injects <think> tags causing 25-46% empty-answer rates without this flag.
See dataset cudabenchmarktest/r9-research-framework/_OLLAMA_INFERENCE_WARNING.md for the full lesson learned.
R8/R9 Five-Bucket… See the full description on the dataset page: https://huggingface.co/datasets/cudabenchmarktest/r8-eval-suite-5bucket.r9-research-framework
R9 Research Framework — Qwen3.5-9B Distillation
⚠️ CRITICAL: READ FIRST — Ollama Inference Flag Required
If you serve any Qwen3.5-derived model from this lineage via Ollama,
you MUST pass "think": false in the /api/chat request body.
curl -X POST http://localhost:11434/api/chat \
-d '{"model": "qwen3.5-9b-r10:q4km", "think": false, "messages": [...], "stream": false}'
Without this flag the model will appear to "loop" and produce empty answers
on 25-46% of requests.… See the full description on the dataset page: https://huggingface.co/datasets/cudabenchmarktest/r9-research-framework.cuda-nsys-training
Qwythos Nsight Systems Profiling Agent Dataset
Multi-turn GPU profiling agent trajectories for fine-tuning Qwythos-9B (and similar tool-calling models) on NVIDIA Nsight Systems (nsys) + CUDA-L1 / KernelBench workloads.
Generated autonomously on an RTX 5090 by the model itself driving real profiling tools for ~33 hours.
Code: ai-hpc/prof-dataset-gen
Stats
Split
Rows
Notes
train
5,884
Accepted episodes (quality ≥ 0.55)
eval
309
5% holdout from accepted… See the full description on the dataset page: https://huggingface.co/datasets/gittensor-model-hub/cuda-nsys-training.r8-thinking-fix-sft
⚠️ CRITICAL: Ollama Inference Flag Required for derived models
If you train or serve any Qwen3.5-9B-derived model from this lineage via Ollama,
you MUST pass "think": false in /api/chat requests for chat / instruction following / tool use.
The qwen3.5 RENDERER auto-injects <think> tags causing 25-46% empty-answer rates without this flag.
See dataset cudabenchmarktest/r9-research-framework/_OLLAMA_INFERENCE_WARNING.md for the full lesson learned.
R8 Thinking-Fix SFT… See the full description on the dataset page: https://huggingface.co/datasets/cudabenchmarktest/r8-thinking-fix-sft.router-bench
router-bench
Open corpus for Gittensor-TinyRouter
by James-Cuda.
Three milestones
Folder
Product goal
Predict / run
Needs
milestone1/
Prompt triage
domain + difficulty
CPU/GPU classifier; no API
milestone2/
Model↔prompt scoring / difficulty routing
which model (from scores or difficulty)
GPU optional; no live API
milestone3/
Full TinyRouter
3 models × 3 roles (Thinker/Worker/Verifier)
GPU + OPENROUTER_API_KEY
You control how many domain labels… See the full description on the dataset page: https://huggingface.co/datasets/James-Cuda/router-bench.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.r7-additive-sft
⚠️ CRITICAL: Ollama Inference Flag Required for derived models
If you train or serve any Qwen3.5-9B-derived model from this lineage via Ollama,
you MUST pass "think": false in /api/chat requests for chat / instruction following / tool use.
The qwen3.5 RENDERER auto-injects <think> tags causing 25-46% empty-answer rates without this flag.
See dataset cudabenchmarktest/r9-research-framework/_OLLAMA_INFERENCE_WARNING.md for the full lesson learned.
R7 Additive SFT… See the full description on the dataset page: https://huggingface.co/datasets/cudabenchmarktest/r7-additive-sft.r8b-tool-sft
⚠️ CRITICAL: Ollama Inference Flag Required for derived models
If you train or serve any Qwen3.5-9B-derived model from this lineage via Ollama,
you MUST pass "think": false in /api/chat requests for chat / instruction following / tool use.
The qwen3.5 RENDERER auto-injects <think> tags causing 25-46% empty-answer rates without this flag.
See dataset cudabenchmarktest/r9-research-framework/_OLLAMA_INFERENCE_WARNING.md for the full lesson learned.
R8b Tool-Deferral +… See the full description on the dataset page: https://huggingface.co/datasets/cudabenchmarktest/r8b-tool-sft.r8-calibration-sft
⚠️ CRITICAL: Ollama Inference Flag Required for derived models
If you train or serve any Qwen3.5-9B-derived model from this lineage via Ollama,
you MUST pass "think": false in /api/chat requests for chat / instruction following / tool use.
The qwen3.5 RENDERER auto-injects <think> tags causing 25-46% empty-answer rates without this flag.
See dataset cudabenchmarktest/r9-research-framework/_OLLAMA_INFERENCE_WARNING.md for the full lesson learned.
R8 Calibration SFT… See the full description on the dataset page: https://huggingface.co/datasets/cudabenchmarktest/r8-calibration-sft.cuda-to-rocm-wavefront-bugs
CUDA → ROCm Wavefront Bug Dataset
170 expert-curated examples of GPU kernel bugs that survive mechanical hipify translation and only manifest on AMD MI300X hardware (gfx942, wavefront-64).
Built for the ROCmPort AI project — a multi-agent pipeline that ports and optimizes CUDA kernels for AMD GPUs.
Why This Dataset Exists
hipify-perl and hipify-clang do a great job of mechanical API renaming (CUDA → HIP). But they cannot detect semantic bugs caused by AMD's larger… See the full description on the dataset page: https://huggingface.co/datasets/tazwarrrr/cuda-to-rocm-wavefront-bugs.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.hygon_cuda_agent
Hygon K100 CUDA Agent Benchmark (Strict v2)
This dataset is re-split with a strict kernel-quality policy.
Strict v2 verified rule
A sample remains in verified only if all conditions hold:
verify_success == true
uses_cuda_extension == true
uses_cudnn_path == false
speedup_vs_baseline <= 3
speedup_vs_compile <= 3
outlier_risk != high
All other verified samples are moved to outlier.
Splits
Split
Count
verified
42
compile_failed
82… See the full description on the dataset page: https://huggingface.co/datasets/mistral0105/hygon_cuda_agent.CUDA-Agent-Ops-6K
CUDA-Agent-Ops-6K
CUDA-Agent-Ops-6K is a curated training dataset for CUDA kernel generation and optimization.
It is released as part of the CUDA-Agent project:
Project Page: https://CUDA-Agent.github.io/
Github Repo: https://github.com/BytedTsinghua-SIA/CUDA-Agent
Dataset Summary
CUDA-Agent-Ops-6K contains 6,000 synthesized operator-level training tasks designed for large-scale agentic RL training. It is intended to provide diverse and executable CUDA-oriented training… See the full description on the dataset page: https://huggingface.co/datasets/kai-celertorai/CUDA-Agent-Ops-6K.CUDA-Agent-Ops-6K
CUDA-Agent-Ops-6K
CUDA-Agent-Ops-6K is a curated training dataset for CUDA kernel generation and optimization.
It is released as part of the CUDA-Agent project:
Project Page: https://CUDA-Agent.github.io/
Github Repo: https://github.com/BytedTsinghua-SIA/CUDA-Agent
Dataset Summary
CUDA-Agent-Ops-6K contains 6,000 synthesized operator-level training tasks designed for large-scale agentic RL training. It is intended to provide diverse and executable CUDA-oriented training… See the full description on the dataset page: https://huggingface.co/datasets/ASKDESC/CUDA-Agent-Ops-6K.