Can I run LFM3-VL-2B on an NVIDIA H200?

Yes

Its published weights (3.84 GB) fit in the 127 GB an NVIDIA H200 can use, with room for ≈ 125K tokens of context.

Full analysis of LFM3-VL-2B Everything that runs on NVIDIA H200 Get the weights on Hugging Face

hf download LiquidAI/LFM3-VL-2B

Files and context on NVIDIA H200

Each file of LFM3-VL-2B (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 H200	Context (16-bit)	Context (8-bit)
As published	3.84 GB	Fits	125K	125K
BF16LiquidAI/LFM3-VL-2B-GGUF	3.47 GB	Fits	125K	125K
F16LiquidAI/LFM3-VL-2B-GGUF	3.47 GB	Fits	125K	125K
Q8_0LiquidAI/LFM3-VL-2B-GGUF	1.85 GB	Fits	125K	125K
Q6_KLiquidAI/LFM3-VL-2B-GGUF	1.43 GB	Fits	125K	125K
LFM3-VL-2B-NVFP4LiquidAI/LFM3-VL-2B-GGUF	1.28 GB	Fits	125K	125K
Q5_K_MLiquidAI/LFM3-VL-2B-GGUF	1.26 GB	Fits	125K	125K
Q4_K_MLiquidAI/LFM3-VL-2B-GGUF	1.10 GB	Fits	125K	125K
Q4_0LiquidAI/LFM3-VL-2B-GGUF	1.05 GB	Fits	125K	125K

LFM3-VL-2B on other hardware

Popular models that run on NVIDIA H200

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.