Starstruk Logo MarkStarstruk

v1.3.30

FoundationVision/VARv1.3.30Jul 27, 2026by github-actions[bot]

AI Summary

This release includes various site improvements, fixes for server origins and speaker labels, and updates to the deployment product hub. It also introduces new runtime downloads and benchmarks for the RTX 4090.

Key Highlights

  • Launches the FunASR deployment product hub.
  • Fixes server to allow trusted browser origins and return speaker labels for requests.
  • Improves MCP discovery and releases version 0.1.2.
  • Adds RTX 4090 community benchmarks.
  • Updates runtime downloads to include llama.cpp binaries for multiple platforms.

New Features

  • Deployment product hub launch
  • Server origin and speaker label fixes
  • MCP discovery improvements
  • RTX 4090 benchmarks
  • Runtime download updates

Full Release Notes

## What's Changed
* ci: guard FunClip v2.1.0 launch pages by @LauraGPT in https://github.com/modelscope/FunASR/pull/3417
* Add MCP Toplist rank badge by @chrstphe in https://github.com/modelscope/FunASR/pull/3418
* fix: limit vLLM VAD segments to 30 seconds by @LauraGPT in https://github.com/modelscope/FunASR/pull/3420
* test: guard donor navigation across public pages by @LauraGPT in https://github.com/modelscope/FunASR/pull/3421
* test: harden donor navigation website contract by @LauraGPT in https://github.com/modelscope/FunASR/pull/3422
* Improve MCP discovery and release 0.1.2 by @LauraGPT in https://github.com/modelscope/FunASR/pull/3423
* feat(site): launch the FunASR deployment product hub by @LauraGPT in https://github.com/modelscope/FunASR/pull/3424
* docs: surface the deployment product hub by @LauraGPT in https://github.com/modelscope/FunASR/pull/3425
* fix(server): allow trusted browser origins by @LauraGPT in https://github.com/modelscope/FunASR/pull/3427
* feat(site): attribute repository conversions by @LauraGPT in https://github.com/modelscope/FunASR/pull/3428
* docs(site): surface Orca SenseVoice desktop integration by @LauraGPT in https://github.com/modelscope/FunASR/pull/3431
* docs: add RTX 4090 Fun-ASR community benchmark by @LauraGPT in https://github.com/modelscope/FunASR/pull/3432
* fix(server): return speaker labels for spk requests by @LauraGPT in https://github.com/modelscope/FunASR/pull/3433
* fix(auto-model): preserve VAD segments on punctuation mismatch by @LauraGPT in https://github.com/modelscope/FunASR/pull/3434
* Decode container-formatted audio bytes before inference by @LauraGPT in https://github.com/modelscope/FunASR/pull/3435
* chore: release FunASR 1.3.30 by @LauraGPT in https://github.com/modelscope/FunASR/pull/3436

## New Contributors
* @chrstphe made their first contribution in https://github.com/modelscope/FunASR/pull/3418

**Full Changelog**: https://github.com/modelscope/FunASR/compare/v1.3.29...v1.3.30

<!-- funasr-runtime-downloads:start -->
## Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: [runtime-llamacpp-v0.1.9](https://github.com/modelscope/FunASR/releases/tag/runtime-llamacpp-v0.1.9).

The same verified runtime assets are attached directly to this Python release so users can find the package and self-contained `llama-funasr-*` binaries in one place.

| Platform | Asset | SHA-256 |
|---|---|---|
| Linux arm64 | [funasr-llamacpp-linux-arm64.tar.gz](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-linux-arm64.tar.gz) | `521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b` |
| Linux x64 AVX2 | [funasr-llamacpp-linux-x64-avx2.tar.gz](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-linux-x64-avx2.tar.gz) | `51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d` |
| Linux x64 Vulkan | [funasr-llamacpp-linux-x64-vulkan.tar.gz](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-linux-x64-vulkan.tar.gz) | `0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7` |
| Linux x64 portable | [funasr-llamacpp-linux-x64.tar.gz](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-linux-x64.tar.gz) | `2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48` |
| macOS arm64 | [funasr-llamacpp-macos-arm64.tar.gz](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-macos-arm64.tar.gz) | `2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363` |
| Windows x64 AVX2 | [funasr-llamacpp-windows-x64-avx2.zip](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-windows-x64-avx2.zip) | `f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c` |
| Windows x64 CUDA | [funasr-llamacpp-windows-x64-cuda.zip](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-windows-x64-cuda.zip) | `3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a` |
| Windows x64 Vulkan | [funasr-llamacpp-windows-x64-vulkan.zip](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-windows-x64-vulkan.zip) | `fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce` |
| Windows x64 portable | [funasr-llamacpp-windows-x64.zip](https://github.com/modelscope/FunASR/releases/download/v1.3.30/funasr-llamacpp-windows-x64.zip) | `6767af74e42c8b928742e12d5995c139636d9482ea151cdbb51f1b7573667772` |

Quick start: download one asset, unpack it, then run the bundled `download-funasr-model.sh <sensevoice|paraformer|nano>` helper and one of `llama-funasr-cli`, `llama-funasr-sensevoice`, or `llama-funasr-paraformer`.

For Python users, install from PyPI:

```bash
python -m pip install -U "funasr==1.3.30"
```
<!-- funasr-runtime-downloads:end -->