In the last 6 months the LocalAI team have created 17 C++ inference...
Richard Palethorpe@jichiep
5 views
Aug 18, 2026
~1 min read
Advertisement
1
In the last 6 months the LocalAI team have created 17 C++ inference backends:
depth-anything.cpp (from the video) — Estimates depth from ordinary images and can produce depth maps or 3D point clouds.
depth-anything.cpp (from the video) — Estimates depth from ordinary images and can produce depth maps or 3D point clouds.
0:33
2
vllm.cpp — Runs and serves large language models efficiently through a lightweight C++ inference engine. Also can create talking llama videos.
0:05
3
trellis2cpp — Generates 3D assets from images using TRELLIS.2. Note it was started by @rms80.
1:06
4
parakeet.cpp — Performs fast local speech recognition using NVIDIA Parakeet models without a Python runtime.
0:13
5
free-splatter.cpp — Reconstructs a 3D scene or object representation from one or more images.
0:27
6
LocalVQE — Cleans speech audio by suppressing noise, echo and reverberation in real time. I also trained this model.
0:14
7
face-detect.cpp — Detects faces and produces face representations suitable for recognition or comparison.
0:09
8
moss-transcribe.cpp — Transcribes audio while identifying who spoke each segment.
0:17
9
moss-tts.cpp — Generates speech locally using MOSS text-to-speech models.
0:13
10
magpie-tts.cpp — Runs Magpie text-to-speech models for local speech generation.
0:11
11
ced.cpp — Classifies environmental and everyday sounds found in audio.
0:13
12
voice-detect.cpp — Detects whether speech or voice activity is present in an audio stream.
0:16
13
locate-anything.cpp — Locates objects or regions in an image from natural-language descriptions.
0:35
14
privacy-filter.cpp — Detects sensitive information such as personal data before text is sent elsewhere.
0:34
15
If you like what we are doing and would like to collaborate then reach out. Also we have started an ADOPTERS.md in LocalAI; adding your organization there is a completely free way of helping out the project.
16
I probably forgot a few, also checkout @LocalAI_API