← Directory
software · Global
TerraMind-1.0-Tokenizer-S2L2A
Hugging Face model ibm-esa-geospatial/TerraMind-1.0-Tokenizer-S2L2A
Claim this listing
Represent TerraMind-1.0-Tokenizer-S2L2A? 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
- Sentinel-2 L2A image encoding to discrete FSQ tokens · Token decoding to image reconstruction (diffusion-based, adjustable timesteps)
- Formats llm-inferred
- Python API (TerraTorch) · PyTorch tensors
- Access docs-verified
- Open licence: Apache-2.0, hosted on Hugging Face; loaded via the TerraTorch registry (no API key stated in DOCS)
Useful for
- Use cases llm-inferred
- Encode Sentinel-2 L2A scenes into discrete tokens for TerraMind generation or fine-tuning workflows · Encode-decode round-trip checks to assess reconstruction fidelity of S-2 imagery · Token-based feature experiments for land cover or change detection prototypes
- Audience llm-inferred
- remote sensing scientists · GIS analysts · geospatial ML engineers
Limitations
- Constraints docs-verified
- Higher diffusion timesteps can add reconstruction detail but may introduce hallucinations · Sentinel-2 L2A only; not designed for other sensors · Geographic coverage, spatial resolution and update cadence not stated in DOCS
- 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; 12166 downloads/30d, 0 likes.
Metadata observed 14/10/2026 · research sweep: manual-2026-10-15 (Hugging Face models)