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

Can I run Spark-X2.5-4B on an NVIDIA H100?

Yes

Its published weights (8.22 GB) fit in the 72.0 GB an NVIDIA H100 can use, with room for ≈ 1M tokens of context.

Full analysis of Spark-X2.5-4B Everything that runs on NVIDIA H100 Get the weights on Hugging Face
hf download XHToken/Spark-X2.5-4B

Files and context on NVIDIA H100

Each file of Spark-X2.5-4B (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 NVIDIA H100 Context (16-bit)Context (8-bit)
As published8.22 GBFits1M1M
Spark-X2.5-4BXHToken/Spark-X2.5-4B-GGUF8.23 GBFits1M1M
Q8_0XHToken/Spark-X2.5-4B-GGUF4.38 GBFits1M1M
Q4_K_MXHToken/Spark-X2.5-4B-GGUF2.60 GBFits1M1M

Spark-X2.5-4B on other hardware

Popular models that run on NVIDIA H100

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.