datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
catalog
Mesh-LLM Catalog
This dataset is the Hugging Face-backed catalog for Mesh-LLM.
The runtime catalog entries live under entries/**/*.json. The Dataset Viewer
uses catalog_rows.jsonl, a flat generated table with one row per model variant.
The catalog deliberately excludes raw blob URLs. Entries should resolve to
Hugging Face repositories and canonical Mesh refs.
details_unit-mesh__autodev-deepseek-6.7b-finetunes-poc
Dataset Card for Evaluation run of unit-mesh/autodev-deepseek-6.7b-finetunes-poc
Dataset automatically created during the evaluation run of model unit-mesh/autodev-deepseek-6.7b-finetunes-poc 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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_unit-mesh__autodev-deepseek-6.7b-finetunes-poc.plugin-catalogagentic-replay-nightlyperformance-history
MeshLLM Performance History
Append-only, machine-readable results from MeshLLM's trusted nightly competitive benchmark. Each immutable run shard contains one normalized row per backend, model, concurrency, hardware/config cohort, and source revision.
The dataset contains performance metrics and content-addressed provenance only. It does not contain prompts, completions, model weights, credentials, local filesystem paths, or raw benchmark logs. GitHub Actions retains the… See the full description on the dataset page: https://huggingface.co/datasets/meshllm/performance-history.bamec66557__VICIOUS_MESH-12B-UNION-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-UNION
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-UNION
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__VICIOUS_MESH-12B-UNION-details.moe-rankings
MoE Rankings
meshllm/moe-rankings is a public dataset of derived Mixture-of-Experts routing metadata for published model artifacts.
The dataset stores ranking artifacts produced by llama-moe-analyze so tools such as mesh-llm can discover expert-hotness rankings for exact model revisions without recomputing them locally.
Purpose
This dataset exists to provide:
immutable MoE expert rankings keyed by exact source model revision
a canonical archive of published ranking… See the full description on the dataset page: https://huggingface.co/datasets/meshllm/moe-rankings.layer-split-outputbamec66557__VICIOUS_MESH-12B-0.X.ver-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-0.X.ver
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-0.X.ver
The dataset is composed of 38 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/open-llm-leaderboard/bamec66557__VICIOUS_MESH-12B-0.X.ver-details.meditsolutions__MedIT-Mesh-3B-Instruct-details
Dataset Card for Evaluation run of meditsolutions/MedIT-Mesh-3B-Instruct
Dataset automatically created during the evaluation run of model meditsolutions/MedIT-Mesh-3B-Instruct
The dataset is composed of 38 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/open-llm-leaderboard/meditsolutions__MedIT-Mesh-3B-Instruct-details.bamec66557__VICIOUS_MESH-12B-ALPHA-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-ALPHA
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-ALPHA
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__VICIOUS_MESH-12B-ALPHA-details.bamec66557__VICIOUS_MESH-12B-OMEGA-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-OMEGA
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-OMEGA
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__VICIOUS_MESH-12B-OMEGA-details.bamec66557__Mistral-Nemo-VICIOUS_MESH-12B-2407-details
Dataset Card for Evaluation run of bamec66557/Mistral-Nemo-VICIOUS_MESH-12B-2407
Dataset automatically created during the evaluation run of model bamec66557/Mistral-Nemo-VICIOUS_MESH-12B-2407
The dataset is composed of 38 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/open-llm-leaderboard/bamec66557__Mistral-Nemo-VICIOUS_MESH-12B-2407-details.bamec66557__VICIOUS_MESH-12B-GAMMA-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-GAMMA
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-GAMMA
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__VICIOUS_MESH-12B-GAMMA-details.bamec66557__VICIOUS_MESH-12B-NEMO-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-NEMO
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-NEMO
The dataset is composed of 38 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/open-llm-leaderboard/bamec66557__VICIOUS_MESH-12B-NEMO-details.bamec66557__VICIOUS_MESH-12B_Razor-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B_Razor
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B_Razor
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__VICIOUS_MESH-12B_Razor-details.bamec66557__VICIOUS_MESH-12B-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B
The dataset is composed of 38 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/bamec66557__VICIOUS_MESH-12B-details.bamec66557__VICIOUS_MESH-12B-0.1v-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-0.1v
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-0.1v
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__VICIOUS_MESH-12B-0.1v-details.bamec66557__VICIOUS_MESH-12B-BETA-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-BETA
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-BETA
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__VICIOUS_MESH-12B-BETA-details.bamec66557__VICIOUS_MESH-12B-DELTA-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-DELTA
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-DELTA
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__VICIOUS_MESH-12B-DELTA-details.bamec66557__VICIOUS_MESH-12B-EPSILON-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-EPSILON
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-EPSILON
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__VICIOUS_MESH-12B-EPSILON-details.bamec66557__VICIOUS_MESH-12B-DIGAMMA-details
Dataset Card for Evaluation run of bamec66557/VICIOUS_MESH-12B-DIGAMMA
Dataset automatically created during the evaluation run of model bamec66557/VICIOUS_MESH-12B-DIGAMMA
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__VICIOUS_MESH-12B-DIGAMMA-details.