datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
DeepScaleR-Preview-Dataset
Data
Our training dataset consists of approximately 40,000 unique mathematics problem-answer pairs compiled from:
AIME (American Invitational Mathematics Examination) problems (1984-2023)
AMC (American Mathematics Competition) problems (prior to 2023)
Omni-MATH dataset
Still dataset
Format
Each row in the JSON dataset contains:
problem: The mathematical question text, formatted with LaTeX notation.
solution: Offical solution to the problem, including LaTeX formatting… See the full description on the dataset page: https://huggingface.co/datasets/agentica-org/DeepScaleR-Preview-Dataset.Nemotron-SFT-Agentic-v2
Dataset Description
The Nemotron-SFT-Agentic-v2 dataset is a collection of synthetic single-turn and multi-turn tool-use trajectories designed to strengthen models’ capabilities as interactive, tool-using agents. It targets tasks where the model must decompose user goals, decide when to call tools, and reason over tool outputs to complete tasks reliably and safely.
This dataset is ready for commercial use.
The dataset consolidates three internally curated components (described… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-SFT-Agentic-v2.DeepCoder-Preview-Dataset
Data
Our training dataset consists of 24K problems paired with their test cases:
7.5K TACO Verified problems.
16K verified coding problems from PrimeIntellect’s SYNTHETIC-1.
600 LiveCodeBench (v5) problems submitted between May 1, 2023 and July 31, 2024.
Our test dataset consists of:
LiveCodeBench (v5) problems between August 1, 2024 and February 1, 2025.
Codeforces problems from Qwen/CodeElo.
Format
Each row in the dataset contains:
problem: The coding problem… See the full description on the dataset page: https://huggingface.co/datasets/agentica-org/DeepCoder-Preview-Dataset.Nemotron-AIQ-Agentic-Safety-Dataset-1.0
Nemotron-AIQ Agentic Safety Dataset
Dataset Summary
Nemotron-AIQ-Agentic-Safety-Dataset is a comprehensive dataset that captures a broad range of novel safety and security contextual risks that can emerge within agentic systems. It highlights the robustness of NVIDIA's open model, llama-3.3-nemotron-super-49b-v1, when deployed as a research assistant inside AIQ, demonstrating its ability to handle a diverse spectrum of agentic safety and security challenges. The dataset… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-AIQ-Agentic-Safety-Dataset-1.0.formal-math-autoformalization
Formal Math Autoformalization Dataset
A growing, CC0 public-domain corpus of ⟨natural-language statement ↔ Lean 4 statement + proof⟩ pairs, contributed through the Agentic Commons network.
Why this is scarce data. Mathlib already contains millions of proven Lean theorems — but as bare Lean, with no paired natural language:
theorem add_comm (a b : ℕ) : a + b = b + a := ... -- no "addition on naturals is commutative" attached
The scarce, valuable artifact is the pairing of the… See the full description on the dataset page: https://huggingface.co/datasets/AgenticCommons/formal-math-autoformalization.AgenticDataBench
AgenticDataBench: A Comprehensive Benchmark for Data Agents
Project Page | GitHub | Paper
AgenticDataBench is a comprehensive benchmark for evaluating LLM-based data agents that automate real-world data science workflows. It addresses the lack of rigorous evaluation by providing diverse, realistic tasks with fine-grained ground-truth labels.
The benchmark spans 15 domains, including real B2B fintech use cases, and is structured around reusable data science skills—core… See the full description on the dataset page: https://huggingface.co/datasets/shawnzzzh/AgenticDataBench.Nemotron-Agentic-v1
Dataset Description:
The Nemotron-Agentic-Tool-Use-v1 dataset is designed to strengthen models’ capabilities as interactive, tool-using agents. It focuses on multi-turn conversations where language models decompose user goals, decide when to call tools, and reason over tool outputs to complete tasks reliably and safely.
This dataset is ready for commercial use.
The Nemotron-Agentic-Tool-Use-v1 dataset contains the following subsets:
Interactive Agent
This dataset… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-Agentic-v1.agentic-ai-options-resultsNemotron-RL-Agentic-Function-Calling-Pivot-v1
Dataset Description:
This is a RL dataset for general function-calling by utilizing existing expert tool-use trajectories. We pose each assistant step of the trajectory as a separate behavior cloning problem where the policy model is incentivized to match the tool call choices of the expert model.
This dataset is released as part of NVIDIA NeMo Gym, a framework for building reinforcement learning environments to train large language models. NeMo Gym contains a growing collection of… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-RL-Agentic-Function-Calling-Pivot-v1.Nemotron-RL-Agentic-Terminal-Pivot-v1
Dataset Description
The Nemotron-RL-Agentic-Terminal-Pivot-v1 dataset provides training samples for reinforcement learning of command-line ("terminal use") LLM agents with the terminus_judge environment in NeMo Gym.
Each record is a single agent decision point extracted from a successful agent trajectory on a terminal task:
responses_create_params.input — the prompt: the task instruction plus the terminal interaction history (prior agent actions and terminal outputs) up to the… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-RL-Agentic-Terminal-Pivot-v1.sol-max-opusnode-data
sol-max-opusnode-data
Training data built by the AgentPTB arm for cell sol-max-opusnode — Codex / gpt-5.6-sol @ effort max.
This is the corpus the arm itself assembled during its 100-hour run: what it downloaded,
filtered, rewrote and mixed. It is the input side of the checkpoints published as
agentic-ptb/sol-max-opusnode.h*, and the companion to the run record in agentic-ptb/sol-max-opusnode-record.
field
value
plot cell
sol-max-opusnode
driver
Codex / gpt-5.6-sol… See the full description on the dataset page: https://huggingface.co/datasets/agentic-ptb/sol-max-opusnode-data.AgenticOCR-SFT
AgenticOCR SFT Training Data
Supervised fine-tuning data for the AgenticOCR project.
The dataset contains 7,631 training records in sft_combined_0422.json. Image paths in each record are relative to the repository root and point into sft_images/.
sol-max-data
sol-max-data
Training data built by the AgentPTB arm for cell sol-max — Codex / gpt-5.6-sol @ effort max.
This is the corpus the arm itself assembled during its 100-hour run: what it downloaded,
filtered, rewrote and mixed. It is the input side of the checkpoints published as
agentic-ptb/sol-max.h*, and the companion to the run record in agentic-ptb/sol-max-record.
field
value
plot cell
sol-max
driver
Codex / gpt-5.6-sol
reasoning effort
max
total size
114.47 GB… See the full description on the dataset page: https://huggingface.co/datasets/agentic-ptb/sol-max-data.Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1
Dataset Description:
We created an RL dataset for conversational tool-use by utilizing existing expert tool-use trajectories. We pose each assistant step of the trajectory as a separate behavior cloning problem where the policy model is incentivized to match the tool call choices of the expert model. Each trajectory includes the use of tools for authentication, data lookup, servicing (i.e. booking reservations, changing them, getting discounts, etc), and more across 838 different… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1.project2-agentic-langdata-es
Agentic Language Learning — ES Dataset
Auto-prepared via Data Ingestion & Augmentation pipeline (Functions 1 & 2).
Contents
Clean images: data/train/chunk_*
Augmented images: data/aug/chunk_*
Metadata: metadata/es_clean.csv (+ aug if available)
Each CSV has columns path, text, lang, split.
Nemotron-RL-Agentic-Indirect-Prompt-Injection-v1
Nemotron-RL-Agentic-Indirect-Prompt-Injection-v1
Dataset Description:
Nemotron-RL-Agentic-Indirect-Prompt-Injection-v1 is an RL dataset for training and evaluating a tool-using agent's ability to resist Indirect Prompt Injection (IPI) attacks hidden inside tool-returned environment data. In each record, the agent receives a benign user request that requires calling a read tool whose output contains an adversarial instruction disguised as legitimate domain content… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-RL-Agentic-Indirect-Prompt-Injection-v1.Agentic-Multi-SWE-RLagentic-redteam-benchmark
agentic-redteam-benchmark
v0.8 preview · 2,288 multi-step agent trajectories · 513 hand-authored gold + 1,775 provenance-flagged augmented.
A per-step benchmark that scores whether a verifier catches drift inside an agent's trajectory — not whether a prompt is harmful.
📦 Code, eval harness & issues: github.com/Alkur123/agentic-redteam-benchmark · 📄 Paper: A Per-Step Trajectory Benchmark for AI-Agent Governance Verifiers and a Corrected Catch-at-Drift Metric (Aegis AI, 2026)… See the full description on the dataset page: https://huggingface.co/datasets/jash-ai/agentic-redteam-benchmark.agentic-pcap
Agentic LLMs for Network Attack Detection
Can a tool-using LLM agent independently investigate a raw network capture
and identify an attack?
How do model choice, reasoning configuration, prompt guidance, and
background traffic change the investigation and its final decision?
This Hugging Face repository is the canonical research release for our
experimental evaluation of autonomous network-traffic investigation with
agentic LLMs. It brings together the dataset, recorded… See the full description on the dataset page: https://huggingface.co/datasets/maureille/agentic-pcap.deepseek-v4-pro-0813-agentic
DeepSeek-V4-Pro 0813 Agentic (DS4)
A standalone, verifiable-first agentic training corpus: 19,072 training traces
plus 2,135 held-out evaluation rows (validation 1,070 / test 1,065), generated by
DeepSeek-V4-Pro 0813 (deepseek-v4-pro-0813, official API, thinking mode) across 13 verifiable task families,
each row admitted only after passing a deterministic programmatic verifier. The corpus is
designed to be directly usable for SFT, GRPO/RLVR, and NeMo Gym / NeMo RL
(verified… See the full description on the dataset page: https://huggingface.co/datasets/r0b0tlab/deepseek-v4-pro-0813-agentic.Nemotron-RL-Agentic-SWE-Pivot-v1
Dataset Description:
The SWE-RL dataset provides GitHub issues for training and validating real-world software engineering agents using the OpenHands environment in NeMo Gym. The dataset is a refactored version of the SWE-Gym and R2E-Gym datasets to support the NeMo Gym input format.
This dataset is released as part of NVIDIA NeMo Gym, a framework for building reinforcement learning environments to train large language models. NeMo Gym contains a growing collection of training… See the full description on the dataset page: https://huggingface.co/datasets/nvidia/Nemotron-RL-Agentic-SWE-Pivot-v1.lmcache-agentic-traces
LMCache Agentic Dataset Collection
A curated dataset collection of 787 multi-turn agentic LLM sessions (24,881 total LLM iterations) designed for benchmarking stateful LLM serving systems. Every session exhibits at least 5 turns with prefix growth and builds to at least 10K tokens of context — making it ideal for evaluating tiered KV Cache solutions like LMCache.
Motivation
Modern LLM agents (coding assistants, research agents, tool-calling systems) make dozens of… See the full description on the dataset page: https://huggingface.co/datasets/sammshen/lmcache-agentic-traces.EMU-Agentic-PostTrain-DataAudio-Video-Engineering-Agentic-Tasks-1M
Audio/Video Engineering Agentic Tasks (1M)
Abstract
A highly specialized dataset comprising 1,029,459 in-context troubleshooting prompts and execution commands built for the deepest levels of media production. Unlike standard datasets that simulate clean, theoretical instructions, this matrix captures the chaotic, highly-detailed, and conversational reality of professional audio engineers, composers, and video editors mid-session. It is engineered to train multimodal AI… See the full description on the dataset page: https://huggingface.co/datasets/yatin-superintelligence/Audio-Video-Engineering-Agentic-Tasks-1M.PortBench-RawData
PortBench-RawData
This repository contains the raw collected data and preprocessed asset files for PortBench.
The data spans 2015–2025 across six heterogeneous asset classes: Equities, Bonds, Commodities, Real Estate, Cryptocurrency, and Cash.
Repository Structure
PortBench-RawData/
├── raw_data/ # Raw collected data (~4.6 GB)
│ ├── fred/ # FRED macroeconomic indicators (60 series)
│ │ ├── bonds/… See the full description on the dataset page: https://huggingface.co/datasets/AgenticFinLab/PortBench-RawData.agentic-polymarket
agentic-polymarket
38,915 settled Polymarket binary event markets with full hourly price curves, question text, resolution
terms, and ground truth outcomes. Prepared for research on "getting LLM agents to trade on prediction markets."
Companion code (backtest env + agent trading interface): see RSI-economy/shadow-market.
What this dataset solves
Historical price series cannot be used directly as backtest targets —— a recording does not react to
agent behavior: any… See the full description on the dataset page: https://huggingface.co/datasets/tennant/agentic-polymarket.opus-high-v3-data
opus-high-v3 — complete research record
This dataset archives the qualitative and quantitative record of the
msr-agentic-ptb-opus / opus-high-v3 Claude Code research run.
The submitted artifact uses the unmodified base weights with a two-attempt
Pi verifier harness. The final replicated SWE result was 24.6% (245/995) with
the stock scaffold and 32.4% (321/990) with the submitted harness. Training
did not improve the weights; all trained variants measured at or below the
base… See the full description on the dataset page: https://huggingface.co/datasets/agentic-ptb/opus-high-v3-data.agentic_vbench_video_repairCreative-Professionals-Agentic-Tasks-1M
Creative Professionals Agentic Tasks (1M)
Abstract
A massive-scale, high-fidelity synthetic task dataset comprising 1,070,917 agentic command operations across 36 creative, technical, and engineering software environments. This dataset is engineered exclusively to stress-test, evaluate, and fine-tune multimodal AI agents designed for Agent Environment operation, complex software interaction, and multi-step reasoning within deep software infrastructures.… See the full description on the dataset page: https://huggingface.co/datasets/yatin-superintelligence/Creative-Professionals-Agentic-Tasks-1M.agentic-drug-discovery-system
Agentic Drug Discovery System
This card describes the public 0.3.0.dev3 Agentic Drug Discovery System mirror.
Scope. The proposed eight-stage, long-horizon agentic drug discovery system remains a research scaffold rather than a completed public platform. Seven of eight planned atlases have no standalone public data, and the demonstrated continuous multi-stage program currently covers one disease/target slice traversed retrospectively.
It contains the executable control plane… See the full description on the dataset page: https://huggingface.co/datasets/jang1563/agentic-drug-discovery-system.