Skip to content
← Directory

software · Global

Prithvi-WxC-1.0-2300M

Hugging Face model ibm-nasa-geospatial/Prithvi-WxC-1.0-2300M (image-feature-extraction)

Claim this listing

Represent Prithvi-WxC-1.0-2300M? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Explore paid sponsorship →

What's in it

Data offered docs-verified
160 MERRA-2 atmospheric variables (pretraining input) · atmospheric state reconstruction from partial information · forecasting / autoregressive rollout · climate downscaling (fine-tuned task) · gravity wave parameterization (fine-tuned task)
Coverage llm-inferred
Global
Formats docs-verified
PyTorch
Access docs-verified
Open model weights under CDLA-Permissive-2.0 licence (Hugging Face); no API key per catalogue metadata

Useful for

Use cases llm-inferred
Regional downscaling of reanalysis fields as meteorological inputs for drainage or flood studies · Gap-filling or reconstruction of missing atmospheric state for rainfall-driven hydrology datasets · Forecast-rollout experiments for short-term meteorological scenarios
Audience llm-inferred
hydrologists · climate risk analysts · drainage engineers with ML pipelines · ML and data scientists

Limitations

Constraints llm-inferred
Not deployed by any Inference Provider, so no hosted API; requires self-hosting or a notebook · Needs a GPU runtime (docs recommend a T4 GPU for the Colab demo) · Trained on MERRA-2 reanalysis, not UK-specific observations; UK suitability needs validation · Model, not a data service: outputs an atmospheric state at one timestamp, with no live feed or river/rainfall gauge data · Only the downscaling and gravity wave tasks are documented as fine-tuned; others need custom work
Licence
cdla-permissive-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; 5468 downloads/30d, 89 likes.

Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face models)