Reposby Glyph
ncnn license, foundation and stars
Model serving and local AI · Custom or mixed (see the LICENSE file) (source-available) · backed by Tencent · 24,022 stars
ncnn is a high-performance neural network inference framework optimized for the mobile platform
License history
LICENSE fileCustom or mixed (see the LICENSE file) No license change recorded since the repository was created in 2017.
Latest release
No GitHub releases published; the project releases elsewhere (tags or its own site).
More in model serving and local ai
OllamaGet up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.Ollama · +684 a week · v0.40.14, Nov 3, 2026
llama.cppLLM inference in C/C++.+611 a week · v0.5.0, Sep 23, 2026
vLLMA high-throughput and memory-efficient inference and serving engine for LLMs.PyTorch Foundation · +562 a week · v0.30.0, Sep 22, 2026
gpt4allGPT4All: Run Local LLMs on Any Device.-5 a weekwhisper.cppPort of OpenAI's Whisper model in C/C++.+160 a week
LocalAILocalAI is the open-source AI engine.+145 a week
JanJan is an open source alternative to ChatGPT that runs 100% offline on your computer.Menlo Research · +114 a week · relicensed May 2025
airllmAirLLM 2.8T LLM inference with single 4GB GPU.+1,139 a week| Project | Stars | |||
|---|---|---|---|---|
 OllamaGet up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.OllamaMIT | MIT | 186k | +684 | v0.40.14Nov 3, 2026 |
 llama.cppLLM inference in C/C++.MIT | MIT | 133k | +611 | v0.5.0Sep 23, 2026 |
 vLLMA high-throughput and memory-efficient inference and serving engine for LLMs.PyTorch FoundationApache 2.0 | Apache 2.0 | 96.6k | +562 | v0.30.0Sep 22, 2026 |
 gpt4allGPT4All: Run Local LLMs on Any Device.MIT | MIT | 77.4k | -5 | - |
| whisper.cppPort of OpenAI's Whisper model in C/C++.MIT | MIT | 55k | +160 | - |
 LocalAILocalAI is the open-source AI engine.MIT | MIT | 50.1k | +145 | - |
 JanJan is an open source alternative to ChatGPT that runs 100% offline on your computer.Menlo ResearchApache 2.0 | Apache 2.0changed May 2025 | 45.2k | +114 | - |
 airllmAirLLM 2.8T LLM inference with single 4GB GPU.Apache 2.0 | Apache 2.0 | 41.9k | +1,139 | - |
About and history
- Repository
- Tencent/ncnn · created Jun 30, 2017
- Language
- C++
- Foundation
- None; run by Tencent
- Company
- Tencent · on Companies · on advisories · on awards · on labs · on Arena · on problems · on socials · on Hotlist
- GitHub detects
- NOASSERTION (a custom or multi-license file; the license above is read from the file itself)
- Stars per week
- 26 (GitHub weekly snapshots)
Weekly snapshots
| Week of | Stars | |
|---|---|---|
| Nov 2, 2026 | 24,022 | |
| Oct 19, 2026 | 23,992 | |
| Oct 12, 2026 | 23,963 | |
| Sep 28, 2026 | 23,906 | |
| Sep 21, 2026 | 23,869 |