← Directory
software · Global
Prithvi-EO-2.0-100M-TL
Hugging Face model ibm-nasa-geospatial/Prithvi-EO-2.0-100M-TL
Claim this listing
Represent Prithvi-EO-2.0-100M-TL? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.
What's in it
- Data offered docs-verified
- pretrained 100M-parameter EO foundation model with temporal and location embeddings · image reconstruction on multitemporal HLS GeoTIFF stacks (inference.py) · fine-tuning via TerraTorch (terratorch>=1.1)
- Resolution docs-verified
- 30 m (pretraining data granularity, NASA HLS V2)
- Formats docs-verified
- GeoTIFF · PyTorch
- Access docs-verified
- Apache-2.0 licence; downloadable from Hugging Face; no API key stated in docs; not deployed by any Inference Provider
Useful for
- Use cases llm-inferred
- Landslide mapping from multitemporal HLS imagery for slope-hazard screening · Land-cover or crop change baseline from HLS time series for habitat or agricultural assessment · Reconstruction of cloud-affected HLS time series before change analysis
- Audience llm-inferred
- GIS analysts · remote sensing specialists · ecologists · hydrologists · ML engineers
Limitations
- Constraints docs-verified
- Lower benchmark performance than larger Prithvi-EO-2.0 variants (lower compute cost) · Not deployed by any Inference Provider; requires local or self-hosted inference · Inputs expected as six HLS bands in reflectance units, in chronological order
- Licence
- apache-2.0
- Authentication
- No API key required (repository README, auto-extracted; check upstream requirements)
- GitHub stars
- Unknown
- Last commit
- Unknown
- Last successful page check
- 14/10/2026
- Page health
- healthy
- Package
- Not recorded
- Verification
- Not independently verified
Provenance & evidence
Verified 2026-10-15 via HF API: exists; 3769 downloads/30d, 2 likes.
Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face models)