Open-weight & owned stack · Quantization efficiency
Quantized K2-Horizon Model Variants Become Available
Quantized GGUF variants of K2-Horizon models ranging from 0.9B to 36B became available while upstream llama.cpp support remained under review.
Read the original at reddit.comOpens the publisher's site in a new tab