datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
power-plant-olmoearth-segmentation
OlmoEarth v1.2-ready power energy dataset
Derived cook 20260830T191500Z from pinned source 35da3d0549b16e15114ded1eb7df3f4a378e9b4f.
The branch uses no train/validation/test split. Every accepted row is tagged all.
Accepted imagery rows: 12453
Quality-unavailable rows: 1
Release mode: partial with documented quality-unavailable exclusions
Clearest-available quality fallbacks: 10
Sentinel-2 L2A: uint16 [12,1,128,128]
Band order: B02,B03,B04,B08,B05,B06,B07,B8A,B11,B12,B01,B09… See the full description on the dataset page: https://huggingface.co/datasets/sarkarghya/power-plant-olmoearth-segmentation.OlmoEarth-v1-Chesapeake-Bay-Nutrient-Loads
OlmoEarth-v1-Chesapeake-Bay-Nutrient-Loads
Annual and monthly water-quality loads (Total Nitrogen, Total Phosphorus, Suspended Sediment) at 121 USGS Nontidal Network monitoring stations across the Chesapeake Bay watershed (2014–2023). Built for use with OlmoEarth-v1-Base embeddings + classical regression, with optional fine-tuning support via quantile-binned classes.
14,166 monthly records · 1,208 annual records · 121 stations · 118 watershed polygons · 6 sub-basins · 4… See the full description on the dataset page: https://huggingface.co/datasets/BAIGroup/OlmoEarth-v1-Chesapeake-Bay-Nutrient-Loads.im3-datacenter-olmoearth-segmentation
OlmoEarth v1.2-ready datacenter energy dataset
Derived cook 20260830T191500Z from pinned source 1f3ee5f334dd53940d088a89db57aed62cc925d5.
The branch uses no train/validation/test split. Every accepted row is tagged all.
Accepted imagery rows: 1417
Quality-unavailable rows: 0
Release mode: partial with documented quality-unavailable exclusions
Clearest-available quality fallbacks: 1
Sentinel-2 L2A: uint16 [12,1,128,128]
Band order: B02,B03,B04,B08,B05,B06,B07,B8A,B11,B12,B01,B09… See the full description on the dataset page: https://huggingface.co/datasets/sarkarghya/im3-datacenter-olmoearth-segmentation.OlmoEarth-v1-Chesapeake-Bay-Nutrient-Loads
OlmoEarth-v1-Chesapeake-Bay-Nutrient-Loads
Annual and monthly water-quality loads (Total Nitrogen, Total Phosphorus, Suspended Sediment) at 121 USGS Nontidal Network monitoring stations across the Chesapeake Bay watershed (2014–2023). Built for use with OlmoEarth-v1-Base embeddings + classical regression, with optional fine-tuning support via quantile-binned classes.
14,166 monthly records · 1,208 annual records · 121 stations · 118 watershed polygons · 6 sub-basins · 4… See the full description on the dataset page: https://huggingface.co/datasets/2imi9/OlmoEarth-v1-Chesapeake-Bay-Nutrient-Loads.wind-and-solar-candidate-olmoearth-segmentation
OlmoEarth v1.2-ready renewable energy dataset
Derived cook 20260830T191500Z from pinned source 2c551f58998cd25554ea679148b21a9c701b51db.
The branch uses no train/validation/test split. Every accepted row is tagged all.
Accepted imagery rows: 12649
Quality-unavailable rows: 4
Release mode: partial with documented quality-unavailable exclusions
Clearest-available quality fallbacks: 4
Sentinel-2 L2A: uint16 [12,1,128,128]
Band order: B02,B03,B04,B08,B05,B06,B07,B8A,B11,B12,B01,B09… See the full description on the dataset page: https://huggingface.co/datasets/sarkarghya/wind-and-solar-candidate-olmoearth-segmentation.OlmoEarth-v1-Potomac-Sewage-Spill-2026
OlmoEarth-v1-Potomac-Sewage-Spill-2026
Validated, time-aligned reference dataset supporting GeoAI tracking of the 2026 Potomac River sewage spill (Glen Echo, MD — 240–300 million gallons released from the 72-inch Potomac Interceptor on January 19, 2026). Provides event timeline, hydrologic context, monitoring station locations, and AOI corridor polygons for AI2's downstream Sentinel-1/2 plume detection.
8 USGS gauges · 846 daily flow records · 4 CBP/CBF stations in AOI · 37 NHD… See the full description on the dataset page: https://huggingface.co/datasets/BAIGroup/OlmoEarth-v1-Potomac-Sewage-Spill-2026.OlmoEarth-v1-Potomac-Sewage-Spill-2026
OlmoEarth-v1-Potomac-Sewage-Spill-2026
Validated, time-aligned reference dataset supporting GeoAI tracking of the 2026 Potomac River sewage spill (Glen Echo, MD — 240–300 million gallons released from the 72-inch Potomac Interceptor on January 19, 2026). Provides event timeline, hydrologic context, monitoring station locations, and AOI corridor polygons for AI2's downstream Sentinel-1/2 plume detection.
8 USGS gauges · 846 daily flow records · 4 CBP/CBF stations in AOI · 37 NHD… See the full description on the dataset page: https://huggingface.co/datasets/2imi9/OlmoEarth-v1-Potomac-Sewage-Spill-2026.key_olmoeOlmoEarth-v1-FT-Karst-Groundwater-Base
OlmoEarth-v1-FT-Karst-Groundwater-Base
Fine-tuning dataset for OlmoEarth-v1-Base to detect karst terrain features in Pennsylvania from Sentinel-1 SAR backscatter and Sentinel-2 optical imagery.
47,020 labeled points · 4 classes · ~50,000 km² · 3 AOI regions
Quick Start
Two formats are provided for the two upload paths:
Option A — OlmoEarth Studio Web UI (studio/import.geojson)
Upload via Studio's "Import your own dataset" feature. Map the properties to your… See the full description on the dataset page: https://huggingface.co/datasets/2imi9/OlmoEarth-v1-FT-Karst-Groundwater-Base.olmoe-0125-1b-7b-preference-mix
OLMoE-1B-7B-0125-Instruct
Note that this collection is licensed under ODC-BY-1.0 license; different licenses apply to subsets of the data. Some portions of the dataset are non-commercial. We present the mixture as a research artifact.
This mix is made up of the following on-policy preference datasets generated using a synthetic data generation pipeline similar to Tulu 3:
Reused prompts from the SFT mix (ai2-adapt-dev/sft_v3.9_used_on_policy_p0_olmoe_1b-7b and… See the full description on the dataset page: https://huggingface.co/datasets/allenai/olmoe-0125-1b-7b-preference-mix.0601_key_cache_dynamic_olmoeOlmoEarth-v1-FT-Karst-Groundwater-Base
OlmoEarth-v1-FT-Karst-Groundwater-Base
Fine-tuning dataset for OlmoEarth-v1-Base to detect karst terrain features in Pennsylvania from Sentinel-1 SAR backscatter and Sentinel-2 optical imagery.
47,020 labeled points · 4 classes · ~50,000 km² · 3 AOI regions
Quick Start
Two formats are provided for the two upload paths:
Option A — OlmoEarth Studio Web UI (studio/import.geojson)
Upload via Studio's "Import your own dataset" feature. Map the properties to your… See the full description on the dataset page: https://huggingface.co/datasets/BAIGroup/OlmoEarth-v1-FT-Karst-Groundwater-Base.0424_key_cache_dynamic_olmoeCore-S2L2A-249k-OlmoEarth-Base
Core-S2L2A-249k-OlmoEarth-Base
This dataset contains pre-computed image embeddings for the Core-S2L2A-249k satellite imagery subset, generated using the OlmoEarth-Base foundation model.
What is OlmoEarth?
OlmoEarth is an Earth-system foundation model trained on Sentinel-2 L2A and 6 derived geospatial maps (OpenStreetMap, WorldCover, SRTM DEM, etc.). It uses a flexible multi-modal vision transformer architecture and excels at capturing spectral and spatial patterns from… See the full description on the dataset page: https://huggingface.co/datasets/Major-TOM/Core-S2L2A-249k-OlmoEarth-Base.0528_key_cache_dynamic_olmoeallenai__OLMoE-1B-7B-0924-Instruct-details
Dataset Card for Evaluation run of allenai/OLMoE-1B-7B-0924-Instruct
Dataset automatically created during the evaluation run of model allenai/OLMoE-1B-7B-0924-Instruct
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/allenai__OLMoE-1B-7B-0924-Instruct-details.allenai__OLMoE-1B-7B-0125-Instruct-details
Dataset Card for Evaluation run of allenai/OLMoE-1B-7B-0125-Instruct
Dataset automatically created during the evaluation run of model allenai/OLMoE-1B-7B-0125-Instruct
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/allenai__OLMoE-1B-7B-0125-Instruct-details.0529_key_cache_dynamic_olmoeallenai__OLMoE-1B-7B-0924-details
Dataset Card for Evaluation run of allenai/OLMoE-1B-7B-0924
Dataset automatically created during the evaluation run of model allenai/OLMoE-1B-7B-0924
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… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard/allenai__OLMoE-1B-7B-0924-details.Wildchat-prefs-280824-olmoetriviaqa_all_OLMoE-1B-7B-0924-Instructentity_all_OLMoE-1B-7B-0924-Instructwildchat_v3.9_unused_p0_olmoe-1b-7bpopqa_all_OLMoE-1B-7B-0924-Instructallenai__OLMoE-1B-7B-0924-Instructkey_cache_olmoeentity-prefix_all_OLMoE-1B-7B-0924-Instruct