Releases · modelscope/FunASR

Release list

FunASR 1.3.27 Latest

Latest

github-actions released this 23 Jul 18:05

v1.3.27

3a5fa06

Weekly Python release focused on OpenAI-compatible serving reliability.

Highlights

verbose_json now reports the language detected by SenseVoice instead of silently defaulting to Chinese.
Fun-ASR-Nano reuses its cached AutoModel after a vLLM fallback, avoiding repeated fallback construction on every request.
Failed vLLM/VAD setup no longer leaves half-initialized engine state when both setup and fallback fail, so a later request can retry.
The OpenAPI metadata and funasr-server startup banner now follow the installed package version.

Install

python -m pip install -U "funasr==1.3.27"

PyPI: https://pypi.org/project/funasr/1.3.27/

The wheel and source distribution are also attached to this GitHub release. Prebuilt llama.cpp / GGUF downloads for Linux, macOS, and Windows are listed below.

Validation

41 server, CLI, documentation, packaging, and runtime contract tests passed.
Wheel and sdist passed twine check.
A clean public PyPI install reports package, CLI, and server version 1.3.27.

Full changelog: v1.3.26...v1.3.27

Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: 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	`521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b`
Linux x64 AVX2	funasr-llamacpp-linux-x64-avx2.tar.gz	`51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d`
Linux x64 Vulkan	funasr-llamacpp-linux-x64-vulkan.tar.gz	`0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7`
Linux x64 portable	funasr-llamacpp-linux-x64.tar.gz	`2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48`
macOS arm64	funasr-llamacpp-macos-arm64.tar.gz	`2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363`
Windows x64 AVX2	funasr-llamacpp-windows-x64-avx2.zip	`f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c`
Windows x64 CUDA	funasr-llamacpp-windows-x64-cuda.zip	`3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a`
Windows x64 Vulkan	funasr-llamacpp-windows-x64-vulkan.zip	`fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce`
Windows x64 portable	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:

python -m pip install -U "funasr==1.3.27"

Assets 13

funasr-1.3.27-py3-none-any.whl

sha256:02ef1e932881b6cbb5530739c7e118f173657c5e9ac2ccb8d2c30b895864ae12

930 KB 2026-07-23T18:06:30Z
funasr-1.3.27.tar.gz

sha256:721601f1eebb74c087b5b85db9ec9d40cf6cfc1c08a3b8a78fe8a01e6715d246

807 KB 2026-07-23T18:06:30Z
funasr-llamacpp-linux-arm64.tar.gz

sha256:521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b

7.25 MB 2026-07-23T22:05:22Z
funasr-llamacpp-linux-x64-avx2.tar.gz

sha256:51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d

7.81 MB 2026-07-23T22:05:22Z
funasr-llamacpp-linux-x64-vulkan.tar.gz

sha256:0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7

19.3 MB 2026-07-23T22:05:22Z
funasr-llamacpp-linux-x64.tar.gz

sha256:2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48

7.27 MB 2026-07-23T22:05:22Z
funasr-llamacpp-macos-arm64.tar.gz

sha256:2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363

6.69 MB 2026-07-23T22:05:22Z
funasr-llamacpp-windows-x64-avx2.zip

sha256:f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c

4.69 MB 2026-07-23T22:05:24Z
funasr-llamacpp-windows-x64-cuda.zip

sha256:3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a

27.1 MB 2026-07-23T22:05:25Z
funasr-llamacpp-windows-x64-vulkan.zip

sha256:fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce

20.8 MB 2026-07-23T22:05:25Z
Source code (zip)

2026-07-23T18:05:36Z
Source code (tar.gz)

2026-07-23T18:05:36Z

FunASR llama.cpp runtime v0.1.9

github-actions released this 23 Jul 11:21

runtime-llamacpp-v0.1.9

9b918d7

Prebuilt self-contained binaries for the FunASR llama.cpp / GGUF runtime: SenseVoice, Paraformer and Fun-ASR-Nano with built-in FSMN-VAD. Download the default quantized model with bash download-funasr-model.sh <sensevoice|paraformer|nano> (the helper requires the Hugging Face CLI: pip install -U huggingface_hub), then run llama-funasr-cli / llama-funasr-sensevoice / llama-funasr-paraformer. Use the default x64 asset for maximum CPU compatibility; use the x64-avx2 asset on CPUs with AVX2/FMA/F16C/BMI2 for higher throughput. The Vulkan assets are linux-x64-vulkan and windows-x64-vulkan; they require a working Vulkan driver/ICD and enable SenseVoiceSmall graph execution with llama-funasr-sensevoice ... --backend vulkan. Build from source with -DGGML_VULKAN=ON to validate platform-specific GPU stacks. The Windows CUDA asset is windows-x64-cuda; it requires an NVIDIA driver compatible with the CUDA Toolkit version configured by the release workflow, targets CUDA architecture 86, and enables SenseVoiceSmall graph execution with llama-funasr-sensevoice ... --backend cuda. Build from source for other GPU architectures. No Python ASR runtime or local build is required. Docs: https://github.com/modelscope/FunASR/blob/runtime-llamacpp-v0.1.9/runtime/llama.cpp/README.md

Assets 11

FunASR MCP server v0.1.1

LauraGPT released this 23 Jul 04:18

mcp-v0.1.1

147a132

FunASR MCP server v0.1.1 is the first fully published, installable release of the official FunASR Model Context Protocol server. It exposes local FunASR/SenseVoice transcription as the transcribe_audio MCP tool over stdio.

Install

The public OCI image is available for Linux amd64:

docker pull ghcr.io/modelscope/funasr-mcp:0.1.1

MCP clients can launch it with a read-only audio directory and a persistent model cache:

docker run --rm -i       --mount type=bind,src=/path/to/audio,dst=/audio,readonly       --mount type=volume,src=funasr-mcp-cache,dst=/root/.cache/modelscope       ghcr.io/modelscope/funasr-mcp:0.1.1

Pass files to the tool as /audio/. The default server uses iic/SenseVoiceSmall; model downloads stay in the named Docker volume between runs.

Registry

Official MCP Registry name: io.github.modelscope/funasr-mcp
OCI package: ghcr.io/modelscope/funasr-mcp:0.1.1
Source and configuration: https://github.com/modelscope/FunASR/tree/mcp-v0.1.1/examples/mcp_server
Container package page: https://github.com/orgs/modelscope/packages/container/package/funasr-mcp

The release pipeline passed metadata/schema validation, container build, MCP initialize/tools-list smoke tests, anonymous GHCR pull verification, GitHub OIDC publication, and official Registry listing verification.

Related downloads

Python/PyPI release: https://github.com/modelscope/FunASR/releases/tag/v1.3.26
Prebuilt llama.cpp/GGUF binaries for Linux, macOS, and Windows: https://github.com/modelscope/FunASR/releases/tag/runtime-llamacpp-v0.1.8
MCP server documentation: https://github.com/modelscope/FunASR/blob/mcp-v0.1.1/examples/mcp_server/README.md

Changes

Publish versioned and latest OCI images to GHCR.
Publish canonical server metadata to the official MCP Registry.
Verify public pull access before Registry publication.
Keep organization package-visibility API limitations from blocking an already-public image.
Preserve a read-only host audio mount and persistent ModelScope model cache.

Assets 2

v1.3.26

github-actions released this 22 Jul 16:18

v1.3.26

f5e8a78

Release focus: OpenAI-compatible ASR deployment

FunASR v1.3.26 is the current Python release for users who want a practical ASR deployment path: install from PyPI, run the OpenAI-compatible transcription API, use WebSocket realtime examples, or pair the release with the prebuilt llama.cpp / GGUF runtime assets below.

Verified deployment evidence collected on 2026-07-23:

PyPI: python -m pip install -U "funasr==1.3.26"
API examples: /v1/audio/transcriptions curl paths are published in examples/openai_api/README.md and examples/openai_api/README_zh.md.
WebSocket examples: serve_realtime_ws.py, client_python.py, and realtime_ws_benchmark.py are documented from docs/vllm_guide.md and docs/vllm_guide_zh.md.
H100 vLLM smoke: Fun-ASR-Nano loaded with vllm 0.19.1, generated 开饭时间早上九点至下午五点。, and processed 1 sample in 1.95s after model load.
Website patrol: funasr.com static page contract passed for 9 pages, including donor, ecosystem, CLI, and llama.cpp/GGUF pages.
Runtime binaries: download the matching runtime-llamacpp-v0.1.9 asset for Linux, macOS, or Windows when you want a self-contained llama-funasr-* binary instead of Python.

What's Changed

docs: require FunASR 1.3.25 for Nano guides by @LauraGPT in #3319
docs: add voice input community integrations by @LauraGPT in #3320
docs: promote merged Chinese LLM resource listing by @LauraGPT in #3322
chore: classify external review wait states by @LauraGPT in #3323
fix: honor server hub for default Nano model by @LauraGPT in #3324
chore: release FunASR 1.3.26 by @LauraGPT in #3325

Full Changelog: v1.3.25...v1.3.26

Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: runtime-llamacpp-v0.1.9.

Use these assets when you want the self-contained llama-funasr-* binaries instead of the Python package. The runtime release is shared by the current Python releases; v1.3.26 itself is the PyPI package release.

Platform	Asset	SHA-256
Linux arm64	funasr-llamacpp-linux-arm64.tar.gz	`521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b`
Linux x64 AVX2	funasr-llamacpp-linux-x64-avx2.tar.gz	`51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d`
Linux x64 Vulkan	funasr-llamacpp-linux-x64-vulkan.tar.gz	`0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7`
Linux x64 portable	funasr-llamacpp-linux-x64.tar.gz	`2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48`
macOS arm64	funasr-llamacpp-macos-arm64.tar.gz	`2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363`
Windows x64 AVX2	funasr-llamacpp-windows-x64-avx2.zip	`f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c`
Windows x64 CUDA	funasr-llamacpp-windows-x64-cuda.zip	`3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a`
Windows x64 Vulkan	funasr-llamacpp-windows-x64-vulkan.zip	`fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce`
Windows x64 portable	funasr-llamacpp-windows-x64.zip	`6767af74e42c8b928742e12d5995c139636d9482ea151cdbb51f1b7573667772`

For Python users, install from PyPI:

python -m pip install -U "funasr==1.3.26"

Contributors

LauraGPT

Assets 2

v1.3.25

github-actions released this 22 Jul 15:13

v1.3.25

006d697

What's Changed

feat: add realtime postprocess hotwords by @LauraGPT in #3317
release: bump funasr to 1.3.25 by @LauraGPT in #3318

Full Changelog: v1.3.24...v1.3.25

Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: runtime-llamacpp-v0.1.9.

Platform	Asset	SHA-256
Linux arm64	funasr-llamacpp-linux-arm64.tar.gz	`521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b`
Linux x64 AVX2	funasr-llamacpp-linux-x64-avx2.tar.gz	`51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d`
Linux x64 Vulkan	funasr-llamacpp-linux-x64-vulkan.tar.gz	`0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7`
Linux x64 portable	funasr-llamacpp-linux-x64.tar.gz	`2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48`
macOS arm64	funasr-llamacpp-macos-arm64.tar.gz	`2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363`
Windows x64 AVX2	funasr-llamacpp-windows-x64-avx2.zip	`f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c`
Windows x64 CUDA	funasr-llamacpp-windows-x64-cuda.zip	`3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a`
Windows x64 Vulkan	funasr-llamacpp-windows-x64-vulkan.zip	`fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce`
Windows x64 portable	funasr-llamacpp-windows-x64.zip	`6767af74e42c8b928742e12d5995c139636d9482ea151cdbb51f1b7573667772`

For Python users, install from PyPI:

python -m pip install -U "funasr==1.3.25"

Contributors

LauraGPT

Assets 2

v1.3.24

github-actions released this 22 Jul 14:49

v1.3.24

622df23

What's Changed

feat(server): add --model-path and --hub options to funasr-server by @liudonghua123 in #3299
fix: client lost the last segment of audio in online mode by @dio-del in #3301
feat(runtime): add GGUF HTTP transcription wrapper by @LauraGPT in #3303
feat(runtime): add Vulkan llama.cpp backend packaging by @LauraGPT in #3304
docs(runtime): mention Vulkan asset in release notes by @LauraGPT in #3305
docs: surface Vulkan llama.cpp runtime in README by @LauraGPT in #3306
docs: clarify realtime partial and hotword behavior by @LauraGPT in #3308
docs: require FunASR 1.3.23 for Nano guides by @LauraGPT in #3310
docs: fix broken markdown links by @LauraGPT in #3311
docs: add canonical Nano finetune Chinese guide path by @LauraGPT in #3312
docs: refresh stale public FunASR links by @LauraGPT in #3313
docs: refresh remaining stale FunASR links by @LauraGPT in #3314
chore: refresh awesome MCP tracking status by @LauraGPT in #3315
release: bump funasr to 1.3.24 by @LauraGPT in #3316

New Contributors

@liudonghua123 made their first contribution in #3299
@dio-del made their first contribution in #3301

Full Changelog: v1.3.23...v1.3.24

Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: runtime-llamacpp-v0.1.9.

Platform	Asset	SHA-256
Linux arm64	funasr-llamacpp-linux-arm64.tar.gz	`521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b`
Linux x64 AVX2	funasr-llamacpp-linux-x64-avx2.tar.gz	`51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d`
Linux x64 Vulkan	funasr-llamacpp-linux-x64-vulkan.tar.gz	`0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7`
Linux x64 portable	funasr-llamacpp-linux-x64.tar.gz	`2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48`
macOS arm64	funasr-llamacpp-macos-arm64.tar.gz	`2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363`
Windows x64 AVX2	funasr-llamacpp-windows-x64-avx2.zip	`f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c`
Windows x64 CUDA	funasr-llamacpp-windows-x64-cuda.zip	`3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a`
Windows x64 Vulkan	funasr-llamacpp-windows-x64-vulkan.zip	`fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce`
Windows x64 portable	funasr-llamacpp-windows-x64.zip	`6767af74e42c8b928742e12d5995c139636d9482ea151cdbb51f1b7573667772`

For Python users, install from PyPI:

python -m pip install -U "funasr==1.3.24"

Contributors

liudonghua123, LauraGPT, and dio-del

Assets 2

v1.3.23

github-actions released this 22 Jul 09:17

v1.3.23

dc9758d

What's Changed

docs: add recording-inbox community integration by @LauraGPT in #3295

Full Changelog: v1.3.22...v1.3.23

Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: runtime-llamacpp-v0.1.9.

Platform	Asset	SHA-256
Linux arm64	funasr-llamacpp-linux-arm64.tar.gz	`521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b`
Linux x64 AVX2	funasr-llamacpp-linux-x64-avx2.tar.gz	`51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d`
Linux x64 Vulkan	funasr-llamacpp-linux-x64-vulkan.tar.gz	`0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7`
Linux x64 portable	funasr-llamacpp-linux-x64.tar.gz	`2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48`
macOS arm64	funasr-llamacpp-macos-arm64.tar.gz	`2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363`
Windows x64 AVX2	funasr-llamacpp-windows-x64-avx2.zip	`f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c`
Windows x64 CUDA	funasr-llamacpp-windows-x64-cuda.zip	`3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a`
Windows x64 Vulkan	funasr-llamacpp-windows-x64-vulkan.zip	`fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce`
Windows x64 portable	funasr-llamacpp-windows-x64.zip	`6767af74e42c8b928742e12d5995c139636d9482ea151cdbb51f1b7573667772`

For Python users, install from PyPI:

python -m pip install -U "funasr==1.3.23"

Contributors

LauraGPT

Assets 2

FunASR llama.cpp runtime v0.1.8

github-actions released this 22 Jul 11:18

runtime-llamacpp-v0.1.8

c7f6edb

Prebuilt self-contained binaries for the FunASR llama.cpp / GGUF runtime: SenseVoice, Paraformer and Fun-ASR-Nano with built-in FSMN-VAD. Download the default quantized model with bash download-funasr-model.sh <sensevoice|paraformer|nano> (the helper requires the Hugging Face CLI: pip install -U huggingface_hub), then run llama-funasr-cli / llama-funasr-sensevoice / llama-funasr-paraformer. Use the default x64 asset for maximum CPU compatibility; use the x64-avx2 asset on CPUs with AVX2/FMA/F16C/BMI2 for higher throughput. The Linux Vulkan asset is linux-x64-vulkan; it requires a working Vulkan driver/ICD and enables SenseVoiceSmall graph execution with llama-funasr-sensevoice ... --backend vulkan. Build from source with -DGGML_VULKAN=ON to validate distro-specific GPU stacks. The Windows CUDA asset is windows-x64-cuda; it requires an NVIDIA driver compatible with the CUDA Toolkit version configured by the release workflow, targets CUDA architecture 86, and enables SenseVoiceSmall graph execution with llama-funasr-sensevoice ... --backend cuda. Build from source for other GPU architectures. No Python ASR runtime or local build is required. Docs: https://github.com/modelscope/FunASR/blob/runtime-llamacpp-v0.1.8/runtime/llama.cpp/README.md

Assets 10

v1.3.22

github-actions released this 19 Jul 11:25

v1.3.22

38e421b

What's Changed

docs: clarify Windows CUDA GPU architecture support by @LauraGPT in #3291
fix(server): segment text-only OpenAI transcription responses by @LauraGPT in #3292
release: prepare funasr 1.3.22 by @LauraGPT in #3293

Full Changelog: v1.3.21...v1.3.22

Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: runtime-llamacpp-v0.1.9.

Platform	Asset	SHA-256
Linux arm64	funasr-llamacpp-linux-arm64.tar.gz	`521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b`
Linux x64 AVX2	funasr-llamacpp-linux-x64-avx2.tar.gz	`51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d`
Linux x64 Vulkan	funasr-llamacpp-linux-x64-vulkan.tar.gz	`0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7`
Linux x64 portable	funasr-llamacpp-linux-x64.tar.gz	`2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48`
macOS arm64	funasr-llamacpp-macos-arm64.tar.gz	`2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363`
Windows x64 AVX2	funasr-llamacpp-windows-x64-avx2.zip	`f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c`
Windows x64 CUDA	funasr-llamacpp-windows-x64-cuda.zip	`3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a`
Windows x64 Vulkan	funasr-llamacpp-windows-x64-vulkan.zip	`fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce`
Windows x64 portable	funasr-llamacpp-windows-x64.zip	`6767af74e42c8b928742e12d5995c139636d9482ea151cdbb51f1b7573667772`

For Python users, install from PyPI:

python -m pip install -U "funasr==1.3.22"

Contributors

LauraGPT

Assets 4

v1.3.21

github-actions released this 19 Jul 10:29

v1.3.21

ac07ca2

What's Changed

fix: lazy load top-level AutoModel exports by @LauraGPT in #3290

Full Changelog: v1.3.20...v1.3.21

Runtime downloads

This Python release pairs with the current prebuilt llama.cpp / GGUF runtime release: runtime-llamacpp-v0.1.9.

Platform	Asset	SHA-256
Linux arm64	funasr-llamacpp-linux-arm64.tar.gz	`521866e75594e56eb5023b65eb1ecf6ab7c3b5069522b71cd33aa37b8406ed4b`
Linux x64 AVX2	funasr-llamacpp-linux-x64-avx2.tar.gz	`51f33822a5191f7963d8ceedba2dd76fe7d810a4388b931b25b8be4f1a8e320d`
Linux x64 Vulkan	funasr-llamacpp-linux-x64-vulkan.tar.gz	`0272856faa30b66831a55dd52c1ea0f2ea04b460251dc84b333d25080afdf6a7`
Linux x64 portable	funasr-llamacpp-linux-x64.tar.gz	`2cd54174a3755f89c11f071dedfb935eff96007617e2e952604d90230ea9eb48`
macOS arm64	funasr-llamacpp-macos-arm64.tar.gz	`2d5786784ad09d8f4def1d942f678728638fe601d00acf0dad7cf094a9328363`
Windows x64 AVX2	funasr-llamacpp-windows-x64-avx2.zip	`f2a1389658e6fb5f5f93c7bad98b5ce100eb4811e0e3c39603e39466773b1b4c`
Windows x64 CUDA	funasr-llamacpp-windows-x64-cuda.zip	`3809fb22ec977b29d86c1ca1d132a16a90019949d7ad2a0262292a5732bf596a`
Windows x64 Vulkan	funasr-llamacpp-windows-x64-vulkan.zip	`fe2988376f911ad88d268edfa95957359015187aa6068076fd9efb2f3cd5d8ce`
Windows x64 portable	funasr-llamacpp-windows-x64.zip	`6767af74e42c8b928742e12d5995c139636d9482ea151cdbb51f1b7573667772`

For Python users, install from PyPI:

python -m pip install -U "funasr==1.3.21"

Contributors

LauraGPT

Assets 2

Uh oh!

Releases: modelscope/FunASR

Release list

FunASR 1.3.27

Highlights

Install

Validation

Runtime downloads

Uh oh!

FunASR llama.cpp runtime v0.1.9

Uh oh!

FunASR MCP server v0.1.1

Install

Registry

Related downloads

Changes

Uh oh!

v1.3.26

Release focus: OpenAI-compatible ASR deployment

What's Changed

Runtime downloads

Contributors

Uh oh!

v1.3.25

What's Changed

Runtime downloads

Contributors

Uh oh!

v1.3.24

What's Changed

New Contributors

Runtime downloads

Contributors

Uh oh!

v1.3.23

What's Changed

Runtime downloads

Contributors

Uh oh!

FunASR llama.cpp runtime v0.1.8

Uh oh!

v1.3.22

What's Changed

Runtime downloads

Contributors

Uh oh!

v1.3.21

What's Changed

Runtime downloads

Contributors

Uh oh!