datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
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.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.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.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.Creative-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.Audio-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.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.agentic-distill-fable-5-sft
Fable-5 SFT — prepared for Qwable fine-tuning
4,659 single-turn pairs from Claude Fable-5 (Anthropic preview model, suspended globally 2026-06-22 under U.S. export-control directives), reformatted into a single-text-column parquet ready for SFTTrainer(dataset_text_field="text") + train_on_responses_only.
Composition:
3,793 rows (81%) end in a <tool_use> block — agentic tool-call patterns
866 rows (19%) end in a pure text response
This is agentic data, not pure reasoning data.… See the full description on the dataset page: https://huggingface.co/datasets/lordx64/agentic-distill-fable-5-sft.Creative-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/rAVEUK/Creative-Professionals-Agentic-Tasks-1M.agentic-coding-trajectories
agentic-coding-trajectories
A unified, tokenized corpus of 15,000 multi-turn agentic-coding sessions (618K turns, 41 turns/session avg) drawn from three publicly-released upstream datasets. Built for benchmarking LLM serving systems on realistic multi-turn coding-agent workloads.
Why this exists
Most LLM serving benchmarks use single-shot prompts. Real coding agents work in long multi-turn loops where each turn appends to a growing prompt. This corpus captures that shape… See the full description on the dataset page: https://huggingface.co/datasets/thoughtworks/agentic-coding-trajectories.H2EPR-Bench
H²EPR-Bench
An Evidence-Traceable Benchmark for Event-Process Reconstruction
H²EPR-Bench asks a demanding question: can a model reconstruct how a complex
real-world event unfolded, rather than merely summarize what happened? Given an
event specification and fixed multi-source evidence, a system produces a
hierarchical heterogeneous Event-Process Graph (EPG) that makes stages,
episodes, participants, actions, outcomes, relations, and evidence support
explicit.… See the full description on the dataset page: https://huggingface.co/datasets/AgenticFinLab/H2EPR-Bench.GitHub-Agentic-PR-Dataset
GitHub Agentic PR Dataset
A large-scale dataset of ~2 million GitHub Pull Requests authored by AI coding agents (Claude Code, Cursor, GitHub Copilot, Devin) and human developers — complete with commits, file-level diffs, patches, and bug-fix classification.
The GitHub Agentic PR Dataset is a research-grade corpus for studying how AI coding agents contribute to real-world open-source software, and how their pull requests compare to those written by humans. It pairs 1,959,649 pull… See the full description on the dataset page: https://huggingface.co/datasets/mabujadallah/GitHub-Agentic-PR-Dataset.Agentic-Chain-of-Thought-Coding-SFT-Dataset
🤖 Agentic Coding CoT Dataset
A high-quality supervised fine-tuning (SFT) dataset for training agentic coding assistants with Chain-of-Thought reasoning capabilities.
📋 Dataset Description
This dataset was created by processing and distilling ~20GB of GitHub crawl data using Minimax-M2 to generate structured, reasoning-rich coding examples. Each sample demonstrates systematic problem-solving with explicit tool usage patterns.
🏗️ Assistant Data Structure… See the full description on the dataset page: https://huggingface.co/datasets/AlicanKiraz0/Agentic-Chain-of-Thought-Coding-SFT-Dataset.Creative-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/kryp1234/Creative-Professionals-Agentic-Tasks-1M.autoinference-agentic-mix-v1
Autoinference Agentic Mix v1
This is a prompt set for the online_agentic serving benchmark. That profile stands
in for long-horizon agent traffic: a large context that grows turn over turn, with
short structured outputs at each step. The usual way to run it uses
generated-shared-prefix, which builds a synthetic shared prefix out of random tokens.
This dataset uses real agent trajectories instead, so the prefix reuse, the context
growth, and the token mix all match what an agent… See the full description on the dataset page: https://huggingface.co/datasets/modal-labs/autoinference-agentic-mix-v1.deepresearchgym-agentic-search-logs
DeepResearchGym Agentic Search Logs
This repository hosts the dataset accompanying the paper “Agentic Search in the Wild” (arXiv: https://arxiv.org/abs/2601.17617).
The dataset contains 14M+ search queries collected via DeepResearchGym (DRGym), an open-source search API designed for DeepResearch-style agentic search. For more background on DRGym, see: https://arxiv.org/abs/2505.19253.
All records have been anonymized and shuffled to prevent re-identification, and we additionally… See the full description on the dataset page: https://huggingface.co/datasets/cx-cmu/deepresearchgym-agentic-search-logs.agentic-llm-pretraining-1.7b
Agentic LLM Pretraining Dataset
A pretraining corpus for small language models (1-3B parameters) optimized for agentic tasks. The corpus emphasizes learning to comprehend language, reason, follow instructions, and use tools over memorizing factual knowledge — the assumption is that domain knowledge will be provided at runtime via RAG. The idea is that this could enable much smaller pretraining corpora by omitting the large volumes of text typically needed to memorize facts.… See the full description on the dataset page: https://huggingface.co/datasets/visionscaper/agentic-llm-pretraining-1.7b.agentic_red_team
Agentic Red Team Tool-Calling Dataset
A multi-turn, tool-calling cybersecurity dataset where each example is a complete agentic trajectory — a realistic sequence of tool calls, tool responses, and reasoning steps that an AI agent would execute during an authorized red team engagement.
Overview
This dataset contains 5,000 agentic tool-calling examples across 20 offensive security sectors. Unlike traditional Q&A datasets, each row is a complete multi-turn trajectory… See the full description on the dataset page: https://huggingface.co/datasets/sh111111111111111/agentic_red_team.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/yuqing1207/Nemotron-AIQ-Agentic-Safety-Dataset-1.0.agentic-rag-redteam-bench
WARNING: HARMFUL CONTENT - RESEARCH USE ONLY
This dataset contains adversarial prompts, jailbreak attacks, toxic outputs, and other explicitly harmful content generated for AI safety research. Samples include prompt injections, social engineering payloads, misinformation, hate speech, instructions for illegal activities, phishing templates, and other dangerous material. All content is synthetic and produced by automated red-teaming pipelines for the sole purpose of evaluating and improving… See the full description on the dataset page: https://huggingface.co/datasets/Fujitsu/agentic-rag-redteam-bench.agentic-safety-gguf
agentic-safety-gguf: Training & Evaluation Datasets
Model: guerilla7/agentic-safety-ggufPaper: (https://arxiv.org/abs/2601.00848)Total: 80,992 examples (80,851 after deduplication)
Overview
Complete training and evaluation datasets for agentic-safety-gguf, a specialized Llama 3.1 8B model for agentic AI security analysis. Supports iterative continuation training methodology (V2→V3→V4) for full reproducibility.
Dataset Files
File
Examples
Size
Purpose… See the full description on the dataset page: https://huggingface.co/datasets/guerilla7/agentic-safety-gguf.TR-HASH-Agentic-SFT-32K-210K
TR-HASH Agentic SFT 32K
Balanced instruction and tool-use SFT data for
AETHORIA-AI/TR-HASH-Tokenizer-32K-Agentic.
The canonical repository name is retained, while its contents replace the former
tool-heavy 21K laboratory corpus.
Composition
Split
General instruction
Tool-aware
Total
Train
182,000
18,000
200,000
Validation
9,000
1,000
10,000
The 9% tool-aware training slice contains tool calls, no-call decisions with
tools present, and final… See the full description on the dataset page: https://huggingface.co/datasets/AETHORIA-AI/TR-HASH-Agentic-SFT-32K-210K.PortBench-QA
PortBench QA Dataset
Dataset Description
6,269 structured question-answer pairs probing correlation-based financial reasoning for multi-asset portfolio management, generated from the PortBench Market Base Dataset.
Task Templates
Template
Task
Complexity
Pairs
T1
Return prediction — direction for next N days
1 (single asset)
1,000
T2
Risk assessment — VaR at given confidence level
1
1,000
T3
Position sizing — given max drawdown… See the full description on the dataset page: https://huggingface.co/datasets/AgenticFinLab/PortBench-QA.opencode-agentic-mini
OpenCode Agentic Dataset
A high-quality training dataset of real-world agentic coding sessions extracted from OpenCode CLI usage. Contains multi-turn conversations with tool calls, reasoning traces, and code editing across 45+ diverse software projects.
Dataset Summary
Split
Examples
Sessions
Train
18,549
~456
Validation
1,001
~51
Total
19,550
507
Estimated tokens: ~153M
Average tokens per example: ~7,800
Data Format
Each… See the full description on the dataset page: https://huggingface.co/datasets/Petrouil/opencode-agentic-mini.autoinference-agentic-mix-v2
Autoinference Agentic Mix v2
300 real SWE-agent trajectories from
TIGER-Lab/SWE-Next-SFT-Trajectories,
expanded into one request per assistant turn. Each row carries the conversation
up to that turn and the model generates the turn. Replaying a trajectory in
turn_index order re-sends a growing prefix, which is how an agent loop
actually hits a prefix cache.
What changed from v1
v1 kept only requests with at least 34k prefix tokens. That cut trajectories
down to… See the full description on the dataset page: https://huggingface.co/datasets/modal-labs/autoinference-agentic-mix-v2.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/DiscoPosse/lmcache-agentic-traces.agentic-task-benchmark
YouMind Agentic Task Benchmark
v0.1.1 · Experimental
YouMindInc/agentic-task-benchmark is a small experimental benchmark for
evaluating creative and research task outcomes against selected references.
This release contains four task descriptions, a standard outcome format, and
an offline text scorer. Task definitions and evaluation protocols may change.
Tasks
Configuration
Task
Status
image_generation
Riso portrait
Defined; reference image supplied… See the full description on the dataset page: https://huggingface.co/datasets/YouMindInc/agentic-task-benchmark.JumpForge-Agentic-SE-3K
JumpForge-Agentic-SE-3K
JumpForge-Agentic-SE-3K is a structured synthetic dataset for training and evaluating
AI software-engineering agents. Its primary target is agent behavior across the software
engineering lifecycle, not raw code generation or memorization of programming-language syntax.
The dataset teaches an agent to:
understand intent and ambiguity before acting;
explore repositories and trace system behavior;
decompose work into reversible steps;
select tools based on… See the full description on the dataset page: https://huggingface.co/datasets/jumplander/JumpForge-Agentic-SE-3K.agentic-reasoning-benchmark
Agentic & Reasoning Benchmark (ARB) – Expanded
Ein synthetischer Benchmark mit 2.550 Fragen und Lösungen, optimiert für die Evaluation von Agentic Capabilities und Reasoning.
Überblick
Eigenschaft
Wert
Anzahl Beispiele
2.550
Kategorien
8
Schwierigkeitsgrade
easy / medium / hard
Formate
CSV + JSON
Reproduzierbarkeit
Generator-Skript (seed=42) enthalten
Lizenz
CC-BY-4.0
Kategorien
Kategorie
Anzahl
Beschreibung… See the full description on the dataset page: https://huggingface.co/datasets/roskosmos19/agentic-reasoning-benchmark.