Can I run K2-Horizon-7B on an NVIDIA B300?

Yes

Its published weights (18.0 GB) fit in the 259 GB an NVIDIA B300 can use, with room for ≈ 512K tokens of context.

Full analysis of K2-Horizon-7B Everything that runs on NVIDIA B300 Get the weights on Hugging Face

hf download IFM/K2-Horizon-7B

Files and context on NVIDIA B300

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.

File	Size	On NVIDIA B300	Context (16-bit)	Context (8-bit)
As published	18.0 GB	Fits	512K	512K
BF16IFM/K2-Horizon-7B-GGUF	18.0 GB	Fits	512K	512K
Q8_0IFM/K2-Horizon-7B-GGUF	9.57 GB	Fits	512K	512K
Q6_KIFM/K2-Horizon-7B-GGUF	7.39 GB	Fits	512K	512K
Q5_K_MIFM/K2-Horizon-7B-GGUF	6.47 GB	Fits	512K	512K
Q5_0IFM/K2-Horizon-7B-GGUF	6.34 GB	Fits	512K	512K
Q4_K_MIFM/K2-Horizon-7B-GGUF	5.59 GB	Fits	512K	512K

K2-Horizon-7B on other hardware

Popular models that run on NVIDIA B300

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.