datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
wildchat-mixed-1k
wildchat-mixed-1k
Real-world chat requests for end-to-end LLM inference benchmarking in fastkernels — Scenario A, the bulk-throughput workload used to saturate continuous batching with a realistic mix of short/long prompts and short/long responses.
What it's for
One dataset that replaces separate prefill-heavy / balanced / decode-heavy splits: its natural length distribution puts prefill-bound and decode-bound requests in the same batch, so a single run yields a… See the full description on the dataset page: https://huggingface.co/datasets/sfc-gh-goliaro/wildchat-mixed-1k.WildChat-curatedAs part of the lock-in hypothesis research project (Qiu et al., 2025), this dataset is transformed from raw WildChat-1M dataset (Zhao et al., 2024) into a structured analysis-ready format through:
Data cleaning by deduplicating users based on IP address co-occurrence and removing templated prompts (i.e. people using the WildChat platform as a free API to do repetitive tasks).
Extracting key concepts from each dialogue using a large language model (Llama-3.1-8B-Instruct), which are then… See the full description on the dataset page: https://huggingface.co/datasets/meno-sh/WildChat-curated.synthid-qwen3-4b-instruct-2507-wildchat
Qwen3-4B SynthID three-arm corpus
This export contains aligned unwatermarked, SynthID key-A, and SynthID key-B
responses from Qwen/Qwen3-4B-Instruct-2507. Matched splits share prompts
and request seeds across configurations; unmatched splits use mutually disjoint
prompt pools.
Export complete for its source work queue: true.
Generation profile
Model revision: cdbee75f17c01a7cc42f958dc650907174af0554
Native model dtype: bfloat16
Maximum generated tokens: 4096… See the full description on the dataset page: https://huggingface.co/datasets/xlr8harder/synthid-qwen3-4b-instruct-2507-wildchat.ro-WildChatThis dataset is a translation of allenai/WildChat using LLMic, a bilingual Romanian-English LLM.
WildChat is a collection of 650K conversations between human users and ChatGPT.
License: ODC-BY
@inproceedings{
zhao2024wildchat,
title={WildChat: 1M Chat{GPT} Interaction Logs in the Wild},
author={Wenting Zhao and Xiang Ren and Jack Hessel and Claire Cardie and Yejin Choi and Yuntian Deng},
booktitle={The Twelfth International Conference on Learning Representations},
year={2024}… See the full description on the dataset page: https://huggingface.co/datasets/faur-ai/ro-WildChat.wildchat-1m-tagged
WildChat 1M with tagging
This dataset replicates the category annotation process introduced in the paper named self-taught evaluator.
This dataset additionally contains three categories, i.e., category, complexity, and length, annotated by Mistral-7B-Instruct-v0.3 for WildChat-1M sessions.
I do my best to follow the technical details explained in the above paper while some difference is inevitably made due to the computational constraint.
The difference and notes for this dataset is… See the full description on the dataset page: https://huggingface.co/datasets/sh0416/wildchat-1m-tagged.wildchat-glm53-format-completions
WildChat format completions
9,975 GLM-5.3 answers across 29 parseable formats. Each answer passed its
contract verifier. Failed answers were resampled with the same prompt until one
passed; no semantic judge or answer repair was used.
This is the final release from a 10,000-prompt run; 25 unfinished prompts were excluded. It stores
answers, format instructions, exact contracts, and pinned
WildChat-4.8M references—not
the source prompts or conversations. All rows are in the train… See the full description on the dataset page: https://huggingface.co/datasets/open-athena/wildchat-glm53-format-completions.wildchat-stratified-sample
WildChat Stratified Sample
Dataset Description
This dataset contains a stratified sample of 263 GPT-4 conversations (347 total turns) from the WildChat dataset. The sample was carefully selected to ensure balanced representation across conversation turn positions and user message lengths.
Dataset Summary
Total Conversations: 263
Total Turns/Rows: 347
Average Turns per Conversation: 1.32
Conversation Length: 1-5 turns (conversations with >5 turns excluded)… See the full description on the dataset page: https://huggingface.co/datasets/Avinaash/wildchat-stratified-sample.wildchat-100k-qwen
WildChat 100k Qwen cleaned
Danish WildChat prompt generations with a cleaned response set. This revision merges regenerated responses for high-refusal target rows, removes high-precision unwanted refusal rows, drops extreme over-length samples, and removes a detected system-prompt leak row.
The dataset keeps the same row schema as the previous synquid/wildchat-100k-qwen upload.
Cleaning summary:
Source rows: 99,983
Kept rows: 99,688
Replaced responses: 3,683
Dropped rows: 295… See the full description on the dataset page: https://huggingface.co/datasets/synquid/wildchat-100k-qwen.