Organization card: TESSERA overview, models, and hero GIF

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tessera-v2-hero-lite.gif filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,10 +1,52 @@
 ---
 title: README
-emoji: 📈
-colorFrom: pink
-colorTo: green
+emoji: 🌍
+colorFrom: green
+colorTo: blue
 sdk: static
 pinned: false
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+<p align="center">
+  <img src="https://huggingface.co/spaces/geotessera/README/resolve/main/tessera-v2-hero-lite.gif"
+       alt="TESSERA — pixel-wise Earth foundation model" width="100%">
+</p>
+
+# GeoTessera
+
+Home of the **TESSERA** family of geospatial foundation models from the
+University of Cambridge ([ucam-eo](https://github.com/ucam-eo)).
+
+TESSERA is a **pixel-wise** Earth-observation foundation model: it encodes a full
+year of **Sentinel-1 and Sentinel-2** observations at each 10 m pixel into a
+compact, task-agnostic embedding for downstream geospatial analysis — land cover,
+crop mapping, biomass and carbon, change detection, and more. No labels, no
+imagery to ship around: just dense annual embeddings you can treat as data.
+
+## Models on this page
+
+| Release | What it is | Repos |
+|---|---|---|
+| **TESSERA v2** (latest) | Four compact pixel students (**Nano 1.07M · Small 7.11M · Medium 21.03M · Large 43.83M**) distilled from a **2B teacher**. Students emit a 128-d **Matryoshka** embedding usable at 16/32/64/128 dims. | [`-2B-N`](https://huggingface.co/geotessera/TESSERA-V-2.0-2B-N) · [`-2B-S`](https://huggingface.co/geotessera/TESSERA-V-2.0-2B-S) · [`-2B-M`](https://huggingface.co/geotessera/TESSERA-V-2.0-2B-M) · [`-2B-L`](https://huggingface.co/geotessera/TESSERA-V-2.0-2B-L) · [`-2B-Teacher`](https://huggingface.co/geotessera/TESSERA-V-2.0-2B-Teacher) |
+| **TESSERA v1.1** | Wider QAT encoder, all-observation inference, per-source normalization; MPC and AWS checkpoints. int8 output. | [`TESSERA-V-1.1`](https://huggingface.co/geotessera/TESSERA-V-1.1) |
+| **TESSERA v1.0** | The original release (QAT / int8 and early fp32). | [`TESSERA-V-1.0`](https://huggingface.co/geotessera/TESSERA-V-1.0) |
+
+Browse everything in the [**TESSERA collection**](https://huggingface.co/collections/geotessera/tessera-6a108c1a6618e155f551c065).
+
+## Two ways to use TESSERA
+
+- **Precomputed embeddings (easiest).** Don't run a model at all — download ready-made
+  global embeddings with the [**`geotessera`**](https://github.com/ucam-eo/geotessera)
+  Python library (`pip install geotessera`), or request coverage for your region.
+- **Run inference yourself.** Generate embeddings from your own Sentinel-1/2 tiles with
+  the code in the [**`tessera`**](https://github.com/ucam-eo/tessera) repository.
+
+## Links
+
+- 📦 Model & inference code: **https://github.com/ucam-eo/tessera**
+- 🐍 Python library for precomputed embeddings: **https://github.com/ucam-eo/geotessera**
+- 🌐 Project website: **https://geotessera.org**
+- 📄 Preprints: [TESSERA v2 — Scaling Pixel-wise Earth Foundation Models](https://arxiv.org/abs/2607.03949) · [TESSERA (v1)](https://arxiv.org/abs/2506.20380)
+
+Model weights are released under **CC0-1.0**; the `geotessera` library is **MIT**.
+Use is additionally governed by the TESSERA [Acceptable Use Policy](https://github.com/ucam-eo/tessera/blob/master/aup.md).