Upload README.md with huggingface_hub

README.md

@@ -4,16 +4,16 @@ pipeline_tag: text-to-speech
 library_name: rlx
 tags:
 - soprano
-- onnx
+- gguf
 - tts
 - rlx
 ---
 
-# Soprano 1.1 ONNX (RLX staging)
+# Soprano 1.1 (RLX GGUF)
 
-Soprano 1.1 KV-cache ONNX export for RLX.
+Packed RLX Soprano 1.1 runnable GGUF (`soprano.gguf`, format `rlx-soprano-gguf-v1`). Not community LM-only Q4 GGUF.
 
-**Kind:** Staging redistrib of an upstream checkpoint for RLX runners.
+**Kind:** RLX-native weight bundle (not a plain upstream mirror).
 
 **Suggested Hub repo:** `eugenehp/soprano`
 
@@ -23,14 +23,14 @@ Soprano 1.1 KV-cache ONNX export for RLX.
 
 ## Contents
 
-Weight tensors and configs in this directory (see file listing on the Hub).
+`soprano.gguf` — pack with `just export-soprano-gguf`.
 
 ## Run with RLX
 
 From [rlx-models](https://github.com/MIT-RLX/rlx-models) with this directory as the model path:
 
 ```bash
-cargo run -p rlx-soprano --release -- --model-dir .
+cargo run -p rlx-soprano --release -- --model-dir . --text "Hi." --output /tmp/soprano.wav
 ```
 
 ## License