Can I run LFM2.5-2.6B on an NVIDIA B300?

Yes

Its published weights (5.39 GB) fit in the 259 GB an NVIDIA B300 can use, with room for ≈ 128K tokens of context.

Full analysis of LFM2.5-2.6B Everything that runs on NVIDIA B300 Get the weights on Hugging Face

hf download LiquidAI/LFM2.5-2.6B

Files and context on NVIDIA B300

Each file of LFM2.5-2.6B (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	5.39 GB	Fits	128K	128K
BF16LiquidAI/LFM2.5-2.6B-GGUF	5.40 GB	Fits	128K	128K
F16LiquidAI/LFM2.5-2.6B-GGUF	5.40 GB	Fits	128K	128K
Q8_0LiquidAI/LFM2.5-2.6B-GGUF	2.87 GB	Fits	128K	128K
Q6_KLiquidAI/LFM2.5-2.6B-GGUF	2.22 GB	Fits	128K	128K
Q5_K_MLiquidAI/LFM2.5-2.6B-GGUF	1.94 GB	Fits	128K	128K
Q4_K_MLiquidAI/LFM2.5-2.6B-GGUF	1.67 GB	Fits	128K	128K
LFM2.5-2.6B-QAD-Q4_0LiquidAI/LFM2.5-2.6B-GGUF	1.59 GB	Fits	128K	128K
Q4_0LiquidAI/LFM2.5-2.6B-GGUF	1.59 GB	Fits	128K	128K

LFM2.5-2.6B 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.