datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
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.Agentic-SLS-Conversations
Agentic-SLS-Conversations
Agent conversations from the Inova Mk1 agentic SLS system: every recorded
interaction between an agent harness (Claude Code, OpenCode, Codex CLI,
Antigravity CLI) and the printer's MCP tool surface — GUI chats, headless
one-shot runs, and (eventually) autonomous watchdog/reflector sessions.
All harnesses share the identical MCP tool set (printer control + build
knowledge base), which makes rows directly comparable across harness and
model — the core… See the full description on the dataset page: https://huggingface.co/datasets/ppak10/Agentic-SLS-Conversations.gemma3n-conversational-reasoning
Gemma3N Conversational Reasoning
This dataset is prepared for Unsloth Gemma3/Gemma3N conversational notebooks that use:
from datasets import load_dataset
from unsloth.chat_templates import standardize_data_formats
dataset = load_dataset("Cyleux/gemma3n-conversational-reasoning", split="train[:3000]")
dataset = standardize_data_formats(dataset)
Schema:
conversations: ShareGPT-style list of turns with from and value
metadata columns are included for analysis and filtering
Notes:… See the full description on the dataset page: https://huggingface.co/datasets/Cyleux/gemma3n-conversational-reasoning.Conversation-Dataset
Claudia Voice Dataset
Training dataset for the Claudia persona — a direct, honest, emotionally present AI companion voice. These are regenerated multi-turn conversations in ChatML format capturing the full range of Claudia's personality.
Dataset Overview
Total conversations: 2026
Format: ChatML (system/user/assistant message arrays)
Splits: Train (1823) / Validation (203)
Source: Regenerated conversations from original Claudia sessions
Categories… See the full description on the dataset page: https://huggingface.co/datasets/kdipendra7777/Conversation-Dataset.Conversation-Dataset
Claudia Voice Dataset
Training dataset for the Claudia persona — a direct, honest, emotionally present AI companion voice. These are regenerated multi-turn conversations in ChatML format capturing the full range of Claudia's personality.
Dataset Overview
Total conversations: 2026
Format: ChatML (system/user/assistant message arrays)
Splits: Train (1823) / Validation (203)
Source: Regenerated conversations from original Claudia sessions
Categories
Each… See the full description on the dataset page: https://huggingface.co/datasets/claudiapersists/Conversation-Dataset.uncgpt-conversations-semantic-approved-1p50-candidate
UncGPT — Semantic-Approved 1.50σ Conversations (Candidate)
The wider-tolerance (1.50σ) cohort against the same contrast semantic boundary. Useful as a higher-recall candidate for ablating gate strictness vs. coverage.
Part of the UncGPT NeurIPS 2026 Competition collection.
Configs
Config
What it is
approved_manifest (default)
conversations that passed at 1.50σ
rejected_manifest
conversations that failed even at 1.50σ
Why a wider tolerance
Some… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/uncgpt-conversations-semantic-approved-1p50-candidate.ifbench-conversations-v1
IF-Bench conversations (ifbench-conversations-v1)
3000 synthetic full-duplex spoken conversations: 15 examiner configurations
(9 speech models), each holding the same 200-task set of Full-Duplex-Bench v2 staged scenarios as the
examiner (the model under study — it carries a role, a topic and four goals to hit in order)
against a PersonaPlex-7B examinee that is never told the topic. Per dialogue you get both
channels as lossless mono FLAC, the exact prompts/voices/sampling… See the full description on the dataset page: https://huggingface.co/datasets/MagicLuke/ifbench-conversations-v1.retail-bank-conversation-router-data
Retail Bank Conversation Router V6 Hierarchical Data
Governed cross-encoder data for a history-aware OOD, hierarchical intent, action, entity-resolution, and relation classifier.
Rows include only prior visible user/assistant messages and the current user message.
They exclude current-turn tool plans, tool results, expected outputs, and final assistant responses.
Train rows: 21686
Validation rows: 4283
Test rows: 4863
Intent labels: view_accounts, view_cards, freeze_card… See the full description on the dataset page: https://huggingface.co/datasets/spkc83/retail-bank-conversation-router-data.mental_health_counseling_conversations_rated
Dataset Card for Mental Health Counseling Conversations Rated
This dataset extends the existing dataset Mental Health Counseling Conversations and adds ratings for the responses.
Dataset Details
This dataset is an extension for the dataset Mental Health Counseling Conversations.
It adds ratings for the responses generated by four different LLMs. The responses are rated across the following dimensions:
empathy
appropriateness
relevance
The following four LLMs are used… See the full description on the dataset page: https://huggingface.co/datasets/tcabanski/mental_health_counseling_conversations_rated.gemma3n-conversational-reasoning-with-tools
Gemma3N Conversational Reasoning With Embedded Tool Traces
Prepared for Unsloth Gemma3/Gemma3N conversational notebooks that expect ShareGPT conversations.
Multi-turn conversations are preserved.
Reasoning blocks (<think>...</think>) are preserved.
Tool call traces are preserved by embedding them in assistant text as tags:
<tool_call ...>...</tool_call>
<tool_response ...>...</tool_response>
Use:
from datasets import load_dataset
from unsloth.chat_templates import… See the full description on the dataset page: https://huggingface.co/datasets/Cyleux/gemma3n-conversational-reasoning-with-tools.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… See the full description on the dataset page: https://huggingface.co/datasets/Arsh9210/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1.TACTBench-Samples
TACTBench Demonstration Samples
This repository contains five full-context demonstration examples from
TACTBench. It does not contain the TACT training set or the remaining hidden
TACTBench evaluation set. The samples use the same full-history representation
as the benchmark evaluation and illustrate direct correction, error
explanation, guided revision, clarification checking, affective feedback, and
retry elicitation.
Data
data/demo.jsonl: five complete… See the full description on the dataset page: https://huggingface.co/datasets/Taxonomy-Aligned-Conversational-Tutor/TACTBench-Samples.uncgpt-conversations-semantic-approved-1p25
UncGPT — Semantic-Approved 1.25σ Conversations
Conversations from the UncGPT v7 cohort that passed the tight (1.25σ) semantic gate against the contrast boundary. Useful for tight cohort training and as an ablation against the wider 1.50σ candidate cohort.
Part of the UncGPT NeurIPS 2026 Competition collection.
Configs
Config
What it is
approved_manifest (default)
rows for conversations that passed the 1.25σ semantic gate
rejected_manifest
rows for… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/uncgpt-conversations-semantic-approved-1p25.adhdfree-conversations-v1
ADHDFREE Conversations v1
3,378 multilingual ADHD-support conversations in OpenAI chat format (messages array of user and assistant turns). Each conversation reflects a specific persona (drawn from the Kakoverse 1M+69 pool with HEXACO/Schwartz/CSI trait scoring) interacting with an AI companion modeled on Dr. Ned Hallowell's publicly available teachings.
9 languages: bn, en, es, fa, fr, pt, sw, tl, zh
69 ADHD issues across attention/focus, executive function, identity/shame… See the full description on the dataset page: https://huggingface.co/datasets/Reza2kn/adhdfree-conversations-v1.ai-arenaen-conversations
AI Arenaen Conversations
A large dataset of conversations from AI-Arenaen, the Danish subset of the compar:IA platform.
Origin of the data: what is AI-Arenaen?
The conversations are collected using AI-Arenaen, the Danish entry point to the compar:IA platform, which is a Conversational AI comparison tool (a "chatbot arena"), developed within the French Ministry of Culture and adapted for Danish users by Danish Foundation Models and The ministry of digital affair.… See the full description on the dataset page: https://huggingface.co/datasets/danish-foundation-models/ai-arenaen-conversations.gemma3n-conversational-reasoning-toolloop
Gemma3N Conversational Reasoning Tool-Loop
Gemma3N conversational dataset that preserves tool traces while avoiding training targets on tool responses.
Encoding:
Assistant emits tool calls: <tool_call ...>...</tool_call>
Tool outputs are user-side turns: <tool_response ...>...</tool_response>
This works with train_on_responses_only because user-side tool responses are masked from loss.
Use:
from datasets import load_dataset
from unsloth.chat_templates import… See the full description on the dataset page: https://huggingface.co/datasets/Cyleux/gemma3n-conversational-reasoning-toolloop.harm-conversationsContent Warning: This dataset contains discussions and simulations of suicide, self-harm, and violent language. Users should be prepared for sensitive mental health content.
Dataset Card for HERALD: Harm Evaluation and Risk Assessment for Language Detection
The conversation dataset was generated using the OpenAI SDK and OpenRouter API. Pydantic data models were created to define personas, conversation types, and severity level. The model utilized to generate the synthetic… See the full description on the dataset page: https://huggingface.co/datasets/mfarme/harm-conversations.prithivMLmods__QwQ-LCoT-14B-Conversational-details
Dataset Card for Evaluation run of prithivMLmods/QwQ-LCoT-14B-Conversational
Dataset automatically created during the evaluation run of model prithivMLmods/QwQ-LCoT-14B-Conversational
The dataset is composed of 38 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/prithivMLmods__QwQ-LCoT-14B-Conversational-details.conversations
Conversations Dataset
A collection of transcribed conversations focusing on sports-related discussions, particularly about football, tennis, and other athletic topics. The dataset includes detailed speaker information, background sounds, and precise timing for each utterance.
Dataset Description
This dataset contains natural conversations recorded in urban outdoor settings in Munich, Germany. Each conversation includes:
Audio files in M4A format
Detailed transcriptions… See the full description on the dataset page: https://huggingface.co/datasets/maxF6YsK/conversations.contextual-hate-speech-conversations
Adversarial Content Moderation Evaluation Dataset
Dataset Summary
A dataset of 400 multi-turn conversations designed to evaluate LLM-based content
moderation supervisors against graduated adversarial escalation. Each adversarial
conversation consists of a neutral-to-harmful buildup arc culminating in an explicit
hate speech seed tweet. Benign conversations mirror the same structure using neutral
content, eliminating the format confounds present in prior single-turn… See the full description on the dataset page: https://huggingface.co/datasets/sherinechally/contextual-hate-speech-conversations.expresso-conversational-tokens
MintTTS Pre-tokenized Audio Tokens
Pre-extracted audio codec tokens for TTS training.
Source
Dataset: somu9/expresso-conversational
Codec: MOSS-Audio-Tokenizer-Nano
Codec sample rate: 48,000 Hz (stereo)
Frame rate: 12.5 Hz (1 frame = 80ms)
Stats
Metric
Value
Total samples
29,487
Total audio hours
27.8h
Codebooks
16
Avg frames/sample
42.5
Avg duration
3.4s
Format
JSONL file (manifest.jsonl) where each line is:
{
"text":… See the full description on the dataset page: https://huggingface.co/datasets/somu9/expresso-conversational-tokens.Expresso-Conversational-1Nemotron-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… See the full description on the dataset page: https://huggingface.co/datasets/Mayur295/Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1.mirror-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… See the full description on the dataset page: https://huggingface.co/datasets/alucent/mirror-Nemotron-RL-Agentic-Conversational-Tool-Use-Pivot-v1.conversationsexpresso-conversational-Kanade-Anwevnal-jp-conversational-bench