Can I run Nanbeige4.3-33B on an RTX 5090?

Yes, with a GGUF version

The published weights are too big, but the Q5_K_M file from AtomicChat/Nanbeige4.3-33B-GGUF (24.7 GB) fits, with room for ≈ 19K tokens of context.

Full analysis of Nanbeige4.3-33B Everything that runs on RTX 5090 Get the Q5_K_M file on Hugging Face

ollama run hf.co/AtomicChat/Nanbeige4.3-33B-GGUF:Q5_K_M

Files and context on RTX 5090

Each file of Nanbeige4.3-33B (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 RTX 5090	Context (16-bit)	Context (8-bit)
As published	66.2 GB	Doesn't fit	—	—
Q8_0AtomicChat/Nanbeige4.3-33B-GGUF	35.2 GB	Doesn't fit	—	—
Q6_KAtomicChat/Nanbeige4.3-33B-GGUF	30.1 GB	Doesn't fit	—	—
Q5_KAtomicChat/Nanbeige4.3-33B-GGUF	27.9 GB	Doesn't fit	4K	8K
Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF	24.7 GB	Tight	19K	38K
Nanbeige4.3-33B-Q5_K_MAtomicChat/Nanbeige4.3-33B-GGUF	23.5 GB	Tight	24K	49K
Nanbeige4.3-33B-AD-Q5_K_M-Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF	22.6 GB	Fits	28K	57K
Q4_K_MAtomicChat/Nanbeige4.3-33B-GGUF	21.2 GB	Fits	35K	70K
IQ4_XSAtomicChat/Nanbeige4.3-33B-GGUF	20.3 GB	Fits	39K	78K
IQ3_SAtomicChat/Nanbeige4.3-33B-GGUF	17.1 GB	Fits	54K	108K
IQ3_XXSAtomicChat/Nanbeige4.3-33B-GGUF	14.9 GB	Fits	64K	127K
IQ2_SAtomicChat/Nanbeige4.3-33B-GGUF	13.7 GB	Fits	69K	138K

Nanbeige4.3-33B on other hardware

Popular models that run on RTX 5090

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.