datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
mergekit-configs
MergeKit-configs: access all Hub architectures and automate your model merging process
This dataset facilitates the search for compatible architectures for model merging with MergeKit, streamlining the automation of high-performance merge searches. It provides a snapshot of the Hub’s configuration state, eliminating the need to manually open configuration files.
import polars as pl
# Login using e.g. `huggingface-cli login` to access this dataset
df =… See the full description on the dataset page: https://huggingface.co/datasets/louisbrulenaudet/mergekit-configs.details_Fischerboot__mergekit-slerp-awrvishdetails_bababababooey__mergekit-slerp-bxtecvo
Dataset Card for Evaluation run of bababababooey/mergekit-slerp-bxtecvo
Dataset automatically created during the evaluation run of model bababababooey/mergekit-slerp-bxtecvo on the Open LLM Leaderboard.
The dataset is composed of 63 configuration, each one coresponding 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_bababababooey__mergekit-slerp-bxtecvo.mergekit-community__VirtuosoSmall-InstructModelStock-details
Dataset Card for Evaluation run of mergekit-community/VirtuosoSmall-InstructModelStock
Dataset automatically created during the evaluation run of model mergekit-community/VirtuosoSmall-InstructModelStock
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mergekit-community__VirtuosoSmall-InstructModelStock-details.DeepMount00__mergekit-ties-okvgjfz-details
Dataset Card for Evaluation run of DeepMount00/mergekit-ties-okvgjfz
Dataset automatically created during the evaluation run of model DeepMount00/mergekit-ties-okvgjfz
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 latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/DeepMount00__mergekit-ties-okvgjfz-details.mergekit-community__JAJUKA-WEWILLNEVERFORGETYOU-3B-details
Dataset Card for Evaluation run of mergekit-community/JAJUKA-WEWILLNEVERFORGETYOU-3B
Dataset automatically created during the evaluation run of model mergekit-community/JAJUKA-WEWILLNEVERFORGETYOU-3B
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mergekit-community__JAJUKA-WEWILLNEVERFORGETYOU-3B-details.mergekit-community__sexeh_time_testing-details
Dataset Card for Evaluation run of mergekit-community/sexeh_time_testing
Dataset automatically created during the evaluation run of model mergekit-community/sexeh_time_testing
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/mergekit-community__sexeh_time_testing-details.details_bababababooey__mergekit-slerp-mntqhzv
Dataset Card for Evaluation run of bababababooey/mergekit-slerp-mntqhzv
Dataset automatically created during the evaluation run of model bababababooey/mergekit-slerp-mntqhzv on the Open LLM Leaderboard.
The dataset is composed of 63 configuration, each one coresponding 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_bababababooey__mergekit-slerp-mntqhzv.details_allknowingroger__mergekit-slerp-zplzqvn
Dataset Card for Evaluation run of allknowingroger/mergekit-slerp-zplzqvn
Dataset automatically created during the evaluation run of model allknowingroger/mergekit-slerp-zplzqvn on the Open LLM Leaderboard.
The dataset is composed of 63 configuration, each one coresponding 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_allknowingroger__mergekit-slerp-zplzqvn.DreadPoor__mergekit-nuslerp-nqzkedi-details
Dataset Card for Evaluation run of DreadPoor/mergekit-nuslerp-nqzkedi
Dataset automatically created during the evaluation run of model DreadPoor/mergekit-nuslerp-nqzkedi
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 latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/DreadPoor__mergekit-nuslerp-nqzkedi-details.baebee__mergekit-model_stock-nzjnheg-details
Dataset Card for Evaluation run of baebee/mergekit-model_stock-nzjnheg
Dataset automatically created during the evaluation run of model baebee/mergekit-model_stock-nzjnheg
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 latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/baebee__mergekit-model_stock-nzjnheg-details.Sakalti__mergekit-della_linear-vmeykci-details
Dataset Card for Evaluation run of Sakalti/mergekit-della_linear-vmeykci
Dataset automatically created during the evaluation run of model Sakalti/mergekit-della_linear-vmeykci
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/Sakalti__mergekit-della_linear-vmeykci-details.baebee__mergekit-ties-fnjenli-detailsmergekit-community__mergekit-ties-ykqemwr-details
Dataset Card for Evaluation run of mergekit-community/mergekit-ties-ykqemwr
Dataset automatically created during the evaluation run of model mergekit-community/mergekit-ties-ykqemwr
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mergekit-community__mergekit-ties-ykqemwr-details.mergekit-community__mergekit-dare_ties-ajgjgea-details
Dataset Card for Evaluation run of mergekit-community/mergekit-dare_ties-ajgjgea
Dataset automatically created during the evaluation run of model mergekit-community/mergekit-dare_ties-ajgjgea
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mergekit-community__mergekit-dare_ties-ajgjgea-details.Sakalti__mergekit-01-details
Dataset Card for Evaluation run of Sakalti/mergekit-01
Dataset automatically created during the evaluation run of model Sakalti/mergekit-01
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 latest results.
An additional… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/Sakalti__mergekit-01-details.master_python_list_mergekit_windows
[!NOTE]
Note: This version works for all archs up until Gemma4. Consider this the "old version" that works for everything up to early 2026.
Master Python List for MergeKit (guaranteed stable on Windows 10)
"The BUG is that after 2-3 paragraphs it just falls apart, broken english, loses all grammar ability and starts repeating words."
It looks like you installed or updated a package that had "aggressive dependencies."
Whatever you installed recently likely said: "I need the… See the full description on the dataset page: https://huggingface.co/datasets/Naphula-Archives/master_python_list_mergekit_windows.mergekit-community__mergekit-della-zgowfmf-details
Dataset Card for Evaluation run of mergekit-community/mergekit-della-zgowfmf
Dataset automatically created during the evaluation run of model mergekit-community/mergekit-della-zgowfmf
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/mergekit-community__mergekit-della-zgowfmf-details.mergekit-community__diabolic6045_ELN-AOC-CAIN-details
Dataset Card for Evaluation run of mergekit-community/diabolic6045_ELN-AOC-CAIN
Dataset automatically created during the evaluation run of model mergekit-community/diabolic6045_ELN-AOC-CAIN
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mergekit-community__diabolic6045_ELN-AOC-CAIN-details.mergekit-community__SuperQwen-2.5-1.5B-details
Dataset Card for Evaluation run of mergekit-community/SuperQwen-2.5-1.5B
Dataset automatically created during the evaluation run of model mergekit-community/SuperQwen-2.5-1.5B
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/mergekit-community__SuperQwen-2.5-1.5B-details.mergekit-community__mergekit-model_stock-azgztvm-details
Dataset Card for Evaluation run of mergekit-community/mergekit-model_stock-azgztvm
Dataset automatically created during the evaluation run of model mergekit-community/mergekit-model_stock-azgztvm
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mergekit-community__mergekit-model_stock-azgztvm-details.bamec66557__mergekit-model_stock-zdaysvi-details
Dataset Card for Evaluation run of bamec66557/mergekit-model_stock-zdaysvi
Dataset automatically created during the evaluation run of model bamec66557/mergekit-model_stock-zdaysvi
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/bamec66557__mergekit-model_stock-zdaysvi-details.bamec66557__mergekit-ties-sinbkow-details
Dataset Card for Evaluation run of bamec66557/mergekit-ties-sinbkow
Dataset automatically created during the evaluation run of model bamec66557/mergekit-ties-sinbkow
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 latest… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/bamec66557__mergekit-ties-sinbkow-details.mergekit-community__mergekit-slerp-fmrazcr-detailsmergekit-community__mergekit-ties-rraxdhv-details
Dataset Card for Evaluation run of mergekit-community/mergekit-ties-rraxdhv
Dataset automatically created during the evaluation run of model mergekit-community/mergekit-ties-rraxdhv
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/mergekit-community__mergekit-ties-rraxdhv-details.master_python_list_mergekit_windows_G4-only
[!NOTE]
Note: This version ONLY works with Gemma4 merges. use the "old version" if you need support for previous archs (Mistral, Llama, etc.)
Master Python List for MergeKit (guaranteed stable on Windows 10 for Gemma 4)
It appears that all recent merge tests in the past 5 days were broken and must be retested. This includes several karcher_stock 24B tests, 8B and 12B merges.
I knew for sure something was wrong once I noticed early terminations with known compatible models that had… See the full description on the dataset page: https://huggingface.co/datasets/Naphula-Archives/master_python_list_mergekit_windows_G4-only.