← Back to all builds
Kimi K3 in Pure C
176KB C99 engine that runs the 2.78 trillion parameter Kimi K3 on a single CPU with 8GB RAM by streaming experts from NVMe. No GPU, no framework.
You can now run a 2.78 trillion parameter model on a CPU with 8.24 GB of RAM 🤯 kimi-k3-in-c is a 176KB pure C99 engine that streams Kimi K3’s experts from disk instead of loading them into memory. no GPU. no CUDA. no framework. 100% Open Source.
