Can I run Qwen3.8-3.6-27B-blend on an NVIDIA H100?

Yes

Its published weights (55.6 GB) fit in the 72.0 GB an NVIDIA H100 can use, with room for ≈ 243K tokens of context.

Full analysis of Qwen3.8-3.6-27B-blend Everything that runs on NVIDIA H100 Get the weights on Hugging Face

hf download JetBrains/Qwen3.8-3.6-27B-blend

Files and context on NVIDIA H100

Each file of Qwen3.8-3.6-27B-blend (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 H100	Context (16-bit)	Context (8-bit)
As published	55.6 GB	Fits	243K	256K
Q5_K_MJetBrains/Qwen3.8-3.6-27B-blend-GGUF	19.5 GB	Fits	256K	256K
Q4_K_MJetBrains/Qwen3.8-3.6-27B-blend-GGUF	16.8 GB	Fits	256K	256K
IQ4_XSJetBrains/Qwen3.8-3.6-27B-blend-GGUF	14.3 GB	Fits	256K	256K
Q3_K_XLJetBrains/Qwen3.8-3.6-27B-blend-GGUF	13.1 GB	Fits	256K	256K
IQ3_SJetBrains/Qwen3.8-3.6-27B-blend-GGUF	12.6 GB	Fits	256K	256K

Qwen3.8-3.6-27B-blend 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.