Ettore Di Giacinto 9684c5dd7e chore(recon): bump pins to cuDNN-conv-capable engines (voice b6e4356, face 6107a24) ...

Adds the opt-in cuDNN implicit-GEMM conv path (VOICEDETECT_GGML_CUDNN /
FACEDETECT_GGML_CUDNN, DEFAULT OFF -> zero build/runtime dep until enabled).
On GPU it kills the im2col-materialization bottleneck and reaches torch-cuDNN
parity on the spill-bound convs: SCRFD detect 14.8->6.4ms (2.3x, ~parity),
WeSpeaker ~parity, ERes2Net beats torch (1.10x); ArcFace/CAM++ neutral (no
spill). Parity exact (SCRFD <=1px, cosine=1.0). To USE it in LocalAI, the CUDA
backend build must enable the flag AND bundle libcudnn - deferred until a
cuDNN-bundled GPU image; flag stays OFF here.

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Assisted-by: Claude:claude-opus-4-8 [Claude Code]

2026-06-24 15:39:42 +00:00

..

acestep-cpp

chore(acestep-cpp): bump pin to ed53caf and adapt wrapper to new API (#9908)

2026-05-20 21:05:32 +00:00

ced

feat(ced): sound-event classification backend (CED audio tagger) (#10425)

2026-06-22 01:00:28 +02:00

cloud-proxy

fix(distributed): self-heal stale 'model not loaded' routing (#10181)

2026-06-05 09:01:36 +02:00

crispasr

fix(crispasr): filter garbage words from parakeet word-level timestamps (#10421)

2026-06-21 17:03:33 +02:00

depth-anything-cpp

feat(gallery): add Depth Anything V2 models + bump native version (#10413)

2026-06-20 14:56:16 +02:00

face-detect

chore(recon): bump pins to cuDNN-conv-capable engines (voice b6e4356, face 6107a24)

2026-06-24 15:39:42 +00:00

llm/llama

feat: add distributed mode (#9124)

2026-03-30 00:47:27 +02:00

local-store

feat(middleware): Model routing, PII filtering, Cloud model proxies (#9802)

2026-05-25 09:28:27 +02:00

localvqe

feat(localvqe/audio): v1.3 release and add spectrograms to audio transform UI (#10113)

2026-05-31 23:56:46 +02:00

locate-anything-cpp

chore: ⬆️ Update mudler/locate-anything.cpp to 92c1682da792c1e8a5dec91acc2be4b02c742ded (#10282)

2026-06-13 09:01:17 +02:00

omnivoice-cpp

chore: ⬆️ Update ServeurpersoCom/omnivoice.cpp to 96d30169afd5e6bb3fd6a0e9be0eb505bfe81fcd (#10408)

2026-06-20 01:36:22 +02:00

opus

feat: add distributed mode (#9124)

2026-03-30 00:47:27 +02:00

parakeet-cpp

chore: ⬆️ Update mudler/parakeet.cpp to db755a78d39f789bb7d4e3935158a9e8105dbe36 (#10393)

2026-06-20 01:37:33 +02:00

piper

fix(package.sh): drop redundant -a and -R

2026-02-05 16:39:38 +01:00

qwen3-tts-cpp

chore: ⬆️ Update ServeurpersoCom/qwentts.cpp to 4536dcdce27c3764a93a06d6bf64026b124962f5 (#10431)

2026-06-22 01:00:10 +02:00

rfdetr-cpp

chore: ⬆️ Update mudler/rf-detr.cpp to 65c0ffcc9a9bc9dae38252f63d0417c9845a6cf7 (#10075)

2026-05-30 00:55:41 +02:00

sam3-cpp

chore: add golangci-lint with new-from-merge-base baseline (#9603)

2026-04-28 22:07:44 +02:00

sherpa-onnx

feat(sherpa-onnx): add Kokoro TTS + multilingual Piper voices (#10309)

2026-06-13 21:27:27 +02:00

silero-vad

fix(package.sh): drop redundant -a and -R

2026-02-05 16:39:38 +01:00

stablediffusion-ggml

chore: ⬆️ Update leejet/stable-diffusion.cpp to b12098f5d09fc83da36e65c784f7bdb16a5a5ebf (#10429)

2026-06-22 00:57:33 +02:00

supertonic

feat(supertonic): add Supertonic ONNX TTS backend (CPU) (#10342)

2026-06-15 16:54:11 +02:00

vibevoice-cpp

fix(darwin): fix vibevoice-cpp build linkage + fail-safe go backend packaging (#10276)

2026-06-12 23:13:50 +02:00

voice-detect

chore(recon): bump pins to cuDNN-conv-capable engines (voice b6e4356, face 6107a24)

2026-06-24 15:39:42 +00:00

voxtral

feat(whisper): honor client cancellation via ggml abort_callback (#9710)

2026-05-08 01:44:47 +02:00

whisper

chore: ⬆️ Update ggml-org/whisper.cpp to 5ed76e9a079962f1c85cfce44edd325c27ef1f97 (#10396)

2026-06-20 01:37:06 +02:00