K2 Horizon 0.9B
LLM360 · 1.1B parameters
The compact dense K2 Horizon reasoning model from IFM's K2/LLM360 program. The released main checkpoint uses multi-domain on-policy distillation and 128K context, and contains 1,078,285,824 BF16 parameters.
Models
252 sources · 1,069 models × 2,256 benchmarks · 113,799 scoresLLM360 · 1.1B parameters
The compact dense K2 Horizon reasoning model from IFM's K2/LLM360 program. The released main checkpoint uses multi-domain on-policy distillation and 128K context, and contains 1,078,285,824 BF16 parameters.