---
license: apache-2.0
base_model: ekwek/Soprano-80M
tags:
  - onnx
---

# Soprano ONNX (KV Cache)

This repository hosts ONNX exports of the Soprano 80M model with KV caching.

## Contents

- `onnx/soprano_backbone_kv.onnx` (backbone with `past_key_values`)
- `onnx/soprano_decoder.onnx` + `onnx/soprano_decoder.onnx.data` (vocoder decoder)
- `/` (tokenizer assets)

## Inference & demo

See the streaming inference code here:
https://github.com/KevinAHM/soprano-web-onnx

Not compatible with WebGPU via onnxruntime-web as of January 2026.

## Upstream

Original project:
https://github.com/ekwek1/soprano