KanTV-AI focus on learning/studying/practising device-AI tech in real application & real scenario on ANY mainstream Android phone equipped with high-end mobile SoC(such as Snapdragon 8Gen3, Snapdragon 8Elite or customized 8Gen3/8Elite......).
The Project KanTV is mainly powered by llama.cpp + whisper.cpp + stable-diffusion.cpp + FFmpeg + opencv-mobile. the core AI engine is driven by ggml, we'd like to make contributions to upstream project llama.cpp.
[2025 May 24]
release v1.6.10[2025 May 02]
release v1.6.8[2025 Mar 11]
submit ggml-hexagon PR to upstream llama.cpp[2025 Feb 12]
create kantv-ai in github and re-launch dev activity of project kantv and ggml-qnn(rename to ggml-hexagon since 2025 March 19)
-
Simple is beautiful
we believe the philosophy of "simple is beautiful" which comes from the great Unix.
-
Make it run, then make it right, then make it fast
-
Explore and have fun!
we believe the philosophy of try crazy ideas, build wild demos, and push the edge of what’s possible(which is one of the core spirits of ggml-way).
-
Universal values
we respect the rule-based (International) order and we respect the IPR, we also respect the fact.
- report issues and contribute PR to KanTV-AI are both greatly welcomed.
- domain tech experts and AI experts are greatly welcomed to join kantv-ai.