datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
gaia2
Gaia2
Paper | Code | Project Page
Dataset Summary
Gaia2 is a benchmark dataset for evaluating AI agent capabilities in simulated environments. The dataset contains 800 scenarios that test agent performance in environments where time flows continuously and events occur dynamically.
The dataset evaluates seven core capabilities: Execution (multi-step planning and state changes), Search (information gathering and synthesis), Adaptability (dynamic response to environmental… See the full description on the dataset page: https://huggingface.co/datasets/meta-agents-research-environments/gaia2.unit4-students-scoresagents-last-exam
Agents Last Exam — Task Card Metadata (v1.1)
A metadata-only release (v1.1) of 152 tasks from the Agents Last Exam (ALE)
benchmark for evaluating computer-use agents on long-horizon professional work.
The Agents Last Exam dataset family
ALE is published as three companion HuggingFace datasets:
Dataset
Contents
Access
Task Card Metadata
One row per task: titles, prompts, taxonomy, input-file descriptors
Open
Task Input Data
The input/ files each task… See the full description on the dataset page: https://huggingface.co/datasets/agents-last-exam/agents-last-exam.trustworthy-biology-agents-traces
Trustworthy Biology Agents — Run Traces
Raw execution traces from 1,329 agent runs across three coding agents on three
biology benchmarks — BiomniBench-DA, BixBench, and CompBioBench. This is the scrubbed
trace bundle for the study in
manu-tej/ai-scientists; the write-up
lives in that repo's RESULTS.md.
The motivating question is not only whether an agent reaches the right answer, but
whether it behaves like a trustworthy analyst when the task is ambiguous,
under-specified, or… See the full description on the dataset page: https://huggingface.co/datasets/amanutej/trustworthy-biology-agents-traces.AgentSearch-V1
Getting Started
The AgentSearch-V1 dataset boasts a comprehensive collection of over one billion embeddings, produced using jina-v2-base. The dataset encompasses more than 50 million high-quality documents and over 1 billion passages, covering a vast range of content from sources such as Arxiv, Wikipedia, Project Gutenberg, and includes carefully filtered Creative Commons (CC) data. Our team is dedicated to continuously expanding and enhancing this corpus to improve the search… See the full description on the dataset page: https://huggingface.co/datasets/SciPhi/AgentSearch-V1.agent-sft-stitch-zh-tts
agent-sft-stitch-zh-tts
Voiced version of voidful/agent-sft-stitch-zh: the STITCH-S spoken chunks synthesized with BlueMagpie-TTS (hung_yi_lee voice), per-utterance loudness-aligned to -23 LUFS, best-of-N + Whisper-CER accepted.
Configs
records (default): one row per agent dialogue — id/source/user/msg (full STITCH-S trajectory) + available_tools + STITCH quality scores + spoken (ordered list of the utterances, each with audio, text, seg_index, cer, accepted… See the full description on the dataset page: https://huggingface.co/datasets/voidful/agent-sft-stitch-zh-tts.course-certificates-of-excellencewave-uiLICENSE
wave-ui-25k
WaveUI-25k
This dataset contains 25k examples of labeled UI elements. It is a subset of a collection of ~80k preprocessed examples assembled from the following sources:
WebUI
RoboFlow
GroundUI-18K
These datasets were preprocessed to have matching schemas and to filter out unwanted examples, such as duplicated, overlapping and low-quality datapoints. We also filtered out many text elements which were not in the main scope of this work.
The WaveUI-25k dataset includes the original… See the full description on the dataset page: https://huggingface.co/datasets/agentsea/wave-ui-25k.unit3-inviteeslectura-agents-data
LectūraAgents Dataset
Overview
This dataset is in support of findings in our paper LectūraAgents: A Multi-Agent Framework for Adaptive Personalized AI-Assisted Learning and Embodied Teaching. LectūaAgents is a hierarchical multi-agent framework that enables end-to-end personalized learning experiences through adaptive embodied teaching. It mirrors a professor–students’ relationship, wherein a ProfessorAgent guides a collaborative team of specialized subordinate… See the full description on the dataset page: https://huggingface.co/datasets/Jaward/lectura-agents-data.tiny-agentsagent-sft-stitch-zh-tts-taste-codec-chat-sample
Gemma 4 E2B Taste-S multi-turn codec SFT
This dataset contains 37,362 complete Traditional Chinese agent
dialogues selected from voidful/agent-sft-stitch-zh-tts. It covers
229,434 synthesized speech segments, approximately
520.5 hours of audio before codec extraction.
Every assistant speech segment is represented without Gemma native audio tags:
<SAY> text_token <a_code> <b_code> ... <p_code> ... </SAY>
The first assistant output starts immediately with <SAY>.
[SOPR]...[EOPR]… See the full description on the dataset page: https://huggingface.co/datasets/voidful/agent-sft-stitch-zh-tts-taste-codec-chat-sample.gaia2-cli
GAIA2 CLI
Benchmark dataset for gaia2-cli, the CLI-based agent evaluation harness.
Schema
Each row has two columns:
Column
Type
Description
scenario_id
string
Unique scenario identifier (e.g. scenario_universe_21_1qgjj6)
scenario
string
Complete scenario as a JSON string
Usage
from datasets import load_dataset
import json
# Load a specific config (160 scenarios)
ds = load_dataset("meta-agents-research-environments/gaia2-cli", "adaptability"… See the full description on the dataset page: https://huggingface.co/datasets/meta-agents-research-environments/gaia2-cli.GroundUI-18K
GroundUI-18K
This dataset is the full GroundUI-18K in AgentStudio. Please note that this dataset is a test set rather than a training set. Therefore, please do not use it for training. More details are provided in the project page.
AgentsNet
AgentsNet
This repository contains the graph instances used in the AgentsNet: Coordination and Collaborative Reasoning in Multi-Agent LLMs paper.
AgentsNet is a new benchmark for multi-agent reasoning, designed to measure the ability of multi-agent systems to collaboratively form strategies for problem-solving, self-organization, and effective communication given a network topology. It draws inspiration from classical problems in distributed systems and graph theory.
Paper:… See the full description on the dataset page: https://huggingface.co/datasets/disco-eth/AgentsNet.CriticBench
Dataset Card for Dataset Name
CriticBench is a comprehensive benchmark designed to assess LLMs' abilities to generate, critique/discriminate and correct reasoning across a variety of tasks. CriticBench encompasses five reasoning domains: mathematical, commonsense, symbolic, coding, and algorithmic. It compiles 15 datasets and incorporates responses from three LLM families.
Dataset Details
Dataset Description
Curated by: THU
Funded by [optional]: [More… See the full description on the dataset page: https://huggingface.co/datasets/llm-agents/CriticBench.Tau2-Bench-Airline-With-Code-Agents
Dataset Card for a Code Agent Version of Tau Bench 2 Airline
Dataset Summary
This dataset includes sample traces and associated metadata from multi-turn interactions between an code agent and AI assistant. The dataset is based on the Airline environment from Tau^2 Bench and contains traces from both the original version and a version made at Snorkel AI using code agents to solve the same tasks (indicator in the version field; details below).
Curated by: Snorkel AI… See the full description on the dataset page: https://huggingface.co/datasets/snorkelai/Tau2-Bench-Airline-With-Code-Agents.agents
Demo Datasets for Weaviate Agents
This datasets repository contains demo datasets for the Weaviate QueryAgent and TransformationAgent.
Query Agent datasets
Brands
Dataset that lists generated brand names, their parent and child brands, what country they are from and so on.
The vectors in this dataset have been created with Snowflake/snowflake-arctic-embed-l-v2.0 via Weaviate Embeddings.
You can load the Brands dataset with vectors, like this:
from datasets… See the full description on the dataset page: https://huggingface.co/datasets/weaviate/agents.insta-150k-v3
InSTA: Towards Internet-Scale Training For Agents
Brandon Trabucco (1) Gunnar Sigurdsson (2) Robinson Piramuthu (2) Ruslan Salakhutdinov (1)
(1) Carnegie Mellon University, Machine Learning Department (2) Amazon
This is a dataset from the authors of the paper Towards Internet-Scale Training For Agents, and contains 150k web navigation tasks to facilitate internet-scale training of LLM agents without relying heavily on human annotations. The dataset is split into… See the full description on the dataset page: https://huggingface.co/datasets/data-for-agents/insta-150k-v3.VFDR-BENCH
VFDR-Bench
VFDR-Bench (Visually Faithful Document Reconstruction Benchmark) evaluates how well a document-to-HTML system preserves a document's text, logical structure, physical layout, and visual styling when converting a page image into HTML — a representation the benchmark calls Fid-HTML.
It was introduced alongside REPLICA, an agentic framework for visually faithful document reconstruction, published at ICDAR 2026 (Oral).
📄 Paper: REPLICA: An Agentic Framework for Visually… See the full description on the dataset page: https://huggingface.co/datasets/replica-agents/VFDR-BENCH.agents_medium_benchmark_3agents_medium_benchmark_2kto_redteaming_data_for_secret_loyaltyTau2-Bench-Verified-Airline-With-Code-Agents
Dataset Card for a Code Agent Version of Tau Bench 2 Airline
Dataset Summary
This dataset includes sample traces and associated metadata from multi-turn interactions between an code agent and AI assistant, along with the original verion of the tasks with more bespoke tools.
The dataset is based on a verified version of the Airline environment from Sierra.ai's Tau^2 Bench with the verified version from Amazon AGI group here.
You can find an earlier version of the dataset… See the full description on the dataset page: https://huggingface.co/datasets/snorkelai/Tau2-Bench-Verified-Airline-With-Code-Agents.ai-code-generation-swe-agents-2026
💻 AI Code Generation, SWE Agents & Program Synthesis Dataset (2026 Edition)
A structured research dataset featuring 3,181 domain-verified research papers and 771 official code repositories focused on Autonomous Software Engineering Agents (SWE-bench), Program Synthesis, DeepSeek-Coder-V2, Qwen2.5-Coder, Test-Driven Code Repair, Self-Healing Software, AST Semantic Modeling, and Formal Logic Verification (2023–2026).
Built with Universal Scientific Engine V17.1 Gold, providing 47… See the full description on the dataset page: https://huggingface.co/datasets/beatsprom/ai-code-generation-swe-agents-2026.agent-sft
voidful/agent-sft
A model-agnostic agent / tool-use SFT dataset in a standard OpenAI-style schema —
train any model on it (Qwen, Llama, Gemma, GPT, …).
Built with the agentds toolkit:
per-source normalization -> group-level dedup (exact + SWE-provenance + MinHash near-dup)
-> heuristic quality stratification. The schema is wire-compatible with
voidful/gemma4-agent-sft
(this run also dedups against it), so the two concatenate cleanly.
Schema
field
type… See the full description on the dataset page: https://huggingface.co/datasets/voidful/agent-sft.AgentSLR
AgentSLR: Priority Pathogens Dataset
Paper
Codebase
Project Website
This dataset accompanies the paper Evaluating AI-based Scientific Knowledge Synthesis with Epidemiological Systematic Reviews. It provides the data component of the AgentSLR evaluation harness. This covers article metadata, human abstract and full text screening labels, and structured human data extractions for epidemiological parameters… See the full description on the dataset page: https://huggingface.co/datasets/OxRML/AgentSLR.insta-150k-v1
InSTA: Towards Internet-Scale Training For Agents
Brandon Trabucco (1) Gunnar Sigurdsson (2) Robinson Piramuthu (2) Ruslan Salakhutdinov (1)
(1) Carnegie Mellon University, Machine Learning Department (2) Amazon
This dataset, presented in the paper Towards Internet-Scale Training For Agents, contains 150k web navigation tasks generated to facilitate Internet-scale training of agents without relying heavily on human annotations. The dataset is split into training and… See the full description on the dataset page: https://huggingface.co/datasets/data-for-agents/insta-150k-v1.agent-skills
Agent Skills Dataset
61,650 agent skills collected from GitHub repositories. Each skill contains name, description, and full markdown content. Useful for skill retrieval and agent training.
Dataset Structure
id: Unique identifier for the skill
name: Skill name
description: Skill description and usage instructions
owner: Repository owner
repo: Repository name
skill_md_path: Path to the skill markdown file
content: Full skill content in markdown format
Usage… See the full description on the dataset page: https://huggingface.co/datasets/LittleDinoC/agent-skills.