Model Anatomy See what's inside open-weight AI models

Can I run K2-Horizon-7B on an RTX 5080?

Yes, with a GGUF version

The published weights are too big, but the Q8_0 file from IFM/K2-Horizon-7B-GGUF (9.57 GB) fits, with room for ≈ 32K tokens of context.

Full analysis of K2-Horizon-7B Everything that runs on RTX 5080 Get the Q8_0 file on Hugging Face
ollama run hf.co/IFM/K2-Horizon-7B-GGUF:Q8_0

Files and context on RTX 5080

Each file of K2-Horizon-7B (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.

FileSizeOn RTX 5080 Context (16-bit)Context (8-bit)
As published18.0 GBDoesn't fit——
BF16IFM/K2-Horizon-7B-GGUF18.0 GBDoesn't fit——
Q8_0IFM/K2-Horizon-7B-GGUF9.57 GBFits32K64K
Q6_KIFM/K2-Horizon-7B-GGUF7.39 GBFits46K93K
Q5_K_MIFM/K2-Horizon-7B-GGUF6.47 GBFits53K105K
Q5_0IFM/K2-Horizon-7B-GGUF6.34 GBFits53K107K
Q4_K_MIFM/K2-Horizon-7B-GGUF5.59 GBFits58K117K

K2-Horizon-7B on other hardware

Popular models that run on RTX 5080

Assumes 90% of the card's memory is usable and at least 4K tokens of context for one conversation, with the context cache in 16-bit unless noted. Fits leaves at least 20% of that memory for context; Tight leaves less.