Document released inference weights and download workflow

README.md

@@ -4,6 +4,38 @@ license: mit
 
 # La-Ribo
 
+## Pretrained weights
+
+The released inference bundle is available in [weights/](https://huggingface.co/GENTEL-Lab/La-Ribo/tree/main/weights).
+
+| File | Purpose |
+|---|---|
+| `weights/la_ribo_base.pt` | Base flow model, exported EMA weights |
+| `weights/la_ribo_tri.pt` | Tri flow model, exported EMA weights |
+| `weights/la_ribo_vae.pt` | Shared autoencoder weights |
+| `weights/latent_stats.pt` | Latent whitening statistics |
+| `weights/coarse_stats.pt` | Scaffold scaling statistics |
+
+The three model files are pure tensor state dictionaries. The normalization files contain only inference-required values and format fields. Base and Tri already contain EMA weights; no `--use_ema` flag is required.
+
+### Download the inference bundle
+
+Run this from the local [La-Ribo inference repository](https://github.com/GENTEL-lab/La-Ribo):
+
+```bash
+uvx --from huggingface_hub hf download GENTEL-Lab/La-Ribo \
+  weights/la_ribo_base.pt \
+  weights/la_ribo_tri.pt \
+  weights/la_ribo_vae.pt \
+  weights/latent_stats.pt \
+  weights/coarse_stats.pt \
+  --local-dir .
+```
+
+Both variants use the shared autoencoder and both statistics files. Choose `config/la-ribo-base.yaml` with `la_ribo_base.pt`, or `config/la-ribo-tri.yaml` with `la_ribo_tri.pt`. Installation and CPU/CUDA inference commands are in the [GitHub README](https://github.com/GENTEL-lab/La-Ribo#readme). The default sampling protocol uses 200 steps and scaffold/latent powers 1.25/2.0.
+
+The data below are optional for inference.
+
 ## Data
 
 The `data/` directory contains 10,631 RNA sequences together with 3D structures predicted by trRosettaRNA2 and per-residue pLDDT scores.