datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
CulturaY
CulturaY: A Large Cleaned Multilingual Dataset of 75 Languages
Dataset Summary
From the team that brought you CulturaX, we present CulturaY, another substantial multilingual dataset of 15TB (uncompressed)/3TB (zstd-compressed) that applies the same dataset cleaning methodology to the HPLT v1.1 dataset.
Please note that HPLT v1.2 has also been released and is an alternative verison with different cleaning methodolgies.
This data was used in part to train our SOTA… See the full description on the dataset page: https://huggingface.co/datasets/Viet-Mistral/CulturaY.mistralai-tekken-toksuite-detokenizedTraining data of the model detokenized in the exact order seen by the model.
The training data is partitioned into 8 chunks (chunk-0 through chunk-7), based on the GPU rank that generated the data. Each chunk contains detokenized text files in JSON Lines format (.jsonl).
qualcomm-interactive-cooking-dataset-ego-mistake-corrections
Qualcomm Interactive Cooking Dataset: Ego Mistake Corrections Benchmark
Description
This dataset contains cooking videos with timestamped instruction and feedback for task guidance.
Each row corresponds to one video and provides aligned lists of utterance text, utterance type, and timestamp.
Dataset Details
Release files:
annotations/annotations.json
videos/*.MP4
Release statistics:
Total videos: 40
Total released annotations: 1,597
Text type counts in… See the full description on the dataset page: https://huggingface.co/datasets/qualcomm/qualcomm-interactive-cooking-dataset-ego-mistake-corrections.mistral_tokenized_2048_fixed_shardslm-eval-results-BarraHome-Mistroll-7B-v2.2-private
Dataset Card for Evaluation run of BarraHome/Mistroll-7B-v2.2
Dataset automatically created during the evaluation run of model BarraHome/Mistroll-7B-v2.2
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 4 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 to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-BarraHome-Mistroll-7B-v2.2-private.sorrel-T-mistral-small-24b-base-seed0-documentsmistral-675b-eval-logs-and-scoresMIST-Bench
✨ MIST Benchmark ✨
A human-annotated benchmark for reasoning under misleading, correct, and irrelevant external signals
Xian Sun1 · Wei Chow2 · Yingshuo Wang3 · Junhao Liu4 · Wei Gao5 · Qing Wu6 · Lingdong Kong2
1 Duke University
·
2 National University of Singapore
·
3 UC Berkeley
·
4 UC Irvine
·
5 Northeastern University
· … See the full description on the dataset page: https://huggingface.co/datasets/worldbench/MIST-Bench.lm-eval-results-chihoonlee10-T3Q-Mistral-Orca-Math-DPO-private
Dataset Card for Evaluation run of chihoonlee10/T3Q-Mistral-Orca-Math-DPO
Dataset automatically created during the evaluation run of model chihoonlee10/T3Q-Mistral-Orca-Math-DPO
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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 to… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-chihoonlee10-T3Q-Mistral-Orca-Math-DPO-private.lm-eval-results-chlee10-T3Q-Merge-Mistral7B-private
Dataset Card for Evaluation run of chlee10/T3Q-Merge-Mistral7B
Dataset automatically created during the evaluation run of model chlee10/T3Q-Merge-Mistral7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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 to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-chlee10-T3Q-Merge-Mistral7B-private.lm-eval-results-nbeerbower-bophades-mistral-truthy-DPO-7B-private
Dataset Card for Evaluation run of nbeerbower/bophades-mistral-truthy-DPO-7B
Dataset automatically created during the evaluation run of model nbeerbower/bophades-mistral-truthy-DPO-7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-nbeerbower-bophades-mistral-truthy-DPO-7B-private.wmt26-mist-sample
Update Log
22 June 2026 (latest) - we updated our data mix because some BELEBELE samples did not have the context. If you downloaded data before 22 June, please download the new version.
16 June 2026 - first version
Summary
The wmt26-mist-sample is a multilingual mix provided by the WMT26 MIST shared task organizers as a starting point for fine-tuning multilingual LLMs. It contains three types of tasks, to cover same-language and cross-lingual comprehension and… See the full description on the dataset page: https://huggingface.co/datasets/pinzhenchen/wmt26-mist-sample.mistralai__Mistral-7B-v0.1-details
Dataset Card for Evaluation run of mistralai/Mistral-7B-v0.1
Dataset automatically created during the evaluation run of model mistralai/Mistral-7B-v0.1
The dataset is composed of 82 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 34 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 to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mistralai__Mistral-7B-v0.1-details.lm-eval-results-teknium-OpenHermes-2.5-Mistral-7B-private
Dataset Card for Evaluation run of teknium/OpenHermes-2.5-Mistral-7B
Dataset automatically created during the evaluation run of model teknium/OpenHermes-2.5-Mistral-7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 6 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 to the latest… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-teknium-OpenHermes-2.5-Mistral-7B-private.lm-eval-results-nbeerbower-bophades-mistral-math-DPO-7B-private
Dataset Card for Evaluation run of nbeerbower/bophades-mistral-math-DPO-7B
Dataset automatically created during the evaluation run of model nbeerbower/bophades-mistral-math-DPO-7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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/nyu-dice-lab/lm-eval-results-nbeerbower-bophades-mistral-math-DPO-7B-private.lm-eval-results-chihoonlee10-T3Q-DPO-Mistral-7B-private
Dataset Card for Evaluation run of chihoonlee10/T3Q-DPO-Mistral-7B
Dataset automatically created during the evaluation run of model chihoonlee10/T3Q-DPO-Mistral-7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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 to the latest… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-chihoonlee10-T3Q-DPO-Mistral-7B-private.mistral_BDIprompt_outputlm-eval-results-chlee10-T3Q-Mistral-Orca-Math-dpo-v2.0-private
Dataset Card for Evaluation run of chlee10/T3Q-Mistral-Orca-Math-dpo-v2.0
Dataset automatically created during the evaluation run of model chlee10/T3Q-Mistral-Orca-Math-dpo-v2.0
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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 to… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-chlee10-T3Q-Mistral-Orca-Math-dpo-v2.0-private.lm-eval-results-pkarypis-mistral-lima-private
Dataset Card for Evaluation run of pkarypis/mistral-lima
Dataset automatically created during the evaluation run of model pkarypis/mistral-lima
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 7 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 to the latest results.
An additional… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-pkarypis-mistral-lima-private.lm-eval-results-nbeerbower-bophades-v2-mistral-7B-private
Dataset Card for Evaluation run of nbeerbower/bophades-v2-mistral-7B
Dataset automatically created during the evaluation run of model nbeerbower/bophades-v2-mistral-7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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 to the latest… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-nbeerbower-bophades-v2-mistral-7B-private.mistral-small-creative-500x
Mistral Small Creative - 500x
This is a non-reasoning dataset created using Mistral Small Creative.
The dataset is meant for creating distilled versions of Mistral Small Creative by fine-tuning already existing open-source LLMs.
This dataset only covers generating short fictional stories.
Meant for testing purposes -> to evaluate if creating a larger dataset would be worth it
Stats
Costs: $ 0.20 (USD)
Total tokens (input + output): 704K
MIST-Train
✨ SCOPE Training Data ✨
Matched preference quartets for Signal-Counterfactual Preference Optimization
Xian Sun1 · Wei Chow2 · Yingshuo Wang3 · Junhao Liu4 · Wei Gao5 · Qing Wu6 · Lingdong Kong2
1 Duke University
·
2 National University of Singapore
·
3 UC Berkeley
·
4 UC Irvine
·
5 Northeastern University
·
6 Nanyang… See the full description on the dataset page: https://huggingface.co/datasets/worldbench/MIST-Train.mistralai__Mixtral-8x22B-v0.1-details
Dataset Card for Evaluation run of mistralai/Mixtral-8x22B-v0.1
Dataset automatically created during the evaluation run of model mistralai/Mixtral-8x22B-v0.1
The dataset is composed of 44 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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 to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mistralai__Mixtral-8x22B-v0.1-details.mistralai__Mistral-Large-Instruct-2411-details
Dataset Card for Evaluation run of mistralai/Mistral-Large-Instruct-2411
Dataset automatically created during the evaluation run of model mistralai/Mistral-Large-Instruct-2411
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 to the… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mistralai__Mistral-Large-Instruct-2411-details.lm-eval-results-chihoonlee10-T3Q-EN-DPO-Mistral-7B-private
Dataset Card for Evaluation run of chihoonlee10/T3Q-EN-DPO-Mistral-7B
Dataset automatically created during the evaluation run of model chihoonlee10/T3Q-EN-DPO-Mistral-7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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 to the… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-chihoonlee10-T3Q-EN-DPO-Mistral-7B-private.lm-eval-results-nbeerbower-slerp-bophades-truthy-math-mistral-7B-private
Dataset Card for Evaluation run of nbeerbower/slerp-bophades-truthy-math-mistral-7B
Dataset automatically created during the evaluation run of model nbeerbower/slerp-bophades-truthy-math-mistral-7B
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 2 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… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-nbeerbower-slerp-bophades-truthy-math-mistral-7B-private.mistralai__Mistral-7B-Instruct-v0.3-details
Dataset Card for Evaluation run of mistralai/Mistral-7B-Instruct-v0.3
Dataset automatically created during the evaluation run of model mistralai/Mistral-7B-Instruct-v0.3
The dataset is composed of 44 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 to the latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mistralai__Mistral-7B-Instruct-v0.3-details.NousResearch__Nous-Hermes-2-Mistral-7B-DPO-details
Dataset Card for Evaluation run of NousResearch/Nous-Hermes-2-Mistral-7B-DPO
Dataset automatically created during the evaluation run of model NousResearch/Nous-Hermes-2-Mistral-7B-DPO
The dataset is composed of 44 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/NousResearch__Nous-Hermes-2-Mistral-7B-DPO-details.lm-eval-results-unaidedelf87777-wizard-mistral-v0.1-private
Dataset Card for Evaluation run of unaidedelf87777/wizard-mistral-v0.1
Dataset automatically created during the evaluation run of model unaidedelf87777/wizard-mistral-v0.1
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 5 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 to the… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-unaidedelf87777-wizard-mistral-v0.1-private.lm-eval-results-mistralai-Mistral-7B-v0.3-private
Dataset Card for Evaluation run of mistralai/Mistral-7B-v0.3
Dataset automatically created during the evaluation run of model mistralai/Mistral-7B-v0.3
The dataset is composed of 62 configuration(s), each one corresponding to one of the evaluated task.
The dataset has been created from 5 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 to the latest results.
An… See the full description on the dataset page: https://huggingface.co/datasets/nyu-dice-lab/lm-eval-results-mistralai-Mistral-7B-v0.3-private.