---
license: apache-2.0
library_name: terratorch
datasets:
  - ibm-esa-geospatial/TerraMesh
tags:
  - Earth Observation
  - TerraMind
  - IBM
  - ESA
---

# TerraMind 1.0 S-2 L2A Tokenizer

TerraMind is the first multimodal any-to-any generative foundation model for Earth Observation jointly developed by IBM, ESA, and Forschungszentrum Jülich.
The model is pre-trained using FSQ-VAE tokens as targets. This tokenizer encodes and decodes Sentinel-2 L2A satellite images for the TerraMind model.

![s2l2a_tokenizer.png](assets%2Fs2l2a_tokenizer.png)

The tokenizer uses FSQ with five dimensions and a codebook size of 15'360 tokens.
The decoding process uses diffusion steps for the reconstruction.
The model was pre-trained for 100 epochs on nine million S-2 L2A images from the TerraMesh dataset.

## Usage

The tokenizer is fully integrated into the fine-tuning toolkit [TerraTorch](https://ibm.github.io/terratorch/).
You can initialize the pre-trained tokenizer with:

```python
from terratorch.registry import FULL_MODEL_REGISTRY
model = FULL_MODEL_REGISTRY.build('terramind_v1_tokenizer_s2l2a', pretrained=True)
```

Once the model is build, it can be used to encode image and decode tokens. 
The number of diffusion steps is defined with `timesteps`. 
Increasing the diffusion steps adds more details to the reconstruction which can also lead to hallucinations. 

```python 
# Encode image
_, _, tokens = model.encode(s2l2a_tensor)
# Decode tokens
reconstruction = model.decode_tokens(tokens, verbose=True, timesteps=10)
# Encode & decode
reconstruction = model(s2l2a_tensor)
```

This tokenizer is automatically loaded with TerraMind generation models like `terramind_v1_base_generate`, see [here](https://huggingface.co/ibm-esa-geospatial/TerraMind-1.0-base#generations) for details.

We provide example code for the tokenizer at https://github.com/IBM/terramind.


## Feedback

If you have feedback or any questions, please start a discussion in this HF repository or submitting an issue to [TerraMind](https://github.com/IBM/terramind) on GitHub.

## Citation

If you use TerraMind in your research, please cite our [TerraMind](https://arxiv.org/abs/2504.11171) pre-print.

```text
@article{jakubik2025terramind,
  title={TerraMind: Large-Scale Generative Multimodality for Earth Observation},
  author={Jakubik, Johannes and Yang, Felix and Blumenstiel, Benedikt and Scheurer, Erik and Sedona, Rocco and Maurogiovanni, Stefano and Bosmans, Jente and Dionelis, Nikolaos and Marsocci, Valerio and Kopp, Niklas and others},
  journal={arXiv preprint arXiv:2504.11171},
  year={2025}
}
```