KleidiAi_support
Pure Q4_0 GGUF conversions built for KleidiAI support: blazing fast on NPU, with even better performance on CPU.
Use anywhere
Standard GGUF files — load them in llama.cpp, Ollama, LM Studio, or anything else that runs GGUF models. No lock-in.
Tuned for FancyAI
These quants are the recommended models in FancyAI (fancyai-os.com), the free and open-source Android app built around on-device NPU/CPU inference.
- Downloads last month
- 2,124
Hardware compatibility
Log In to add your hardware
4-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support