Can I run Qwen3.6-27B-MTP-GGUF on an RTX 5090?

Yes, just

Its UD-Q6_K_XL file (26.0 GB) fits in the 28.8 GB an RTX 5090 can use, with room for ≈ 39K tokens of context.

Full analysis of Qwen3.6-27B-MTP-GGUF Everything that runs on RTX 5090 Get the UD-Q6_K_XL file on Hugging Face

ollama run hf.co/unsloth/Qwen3.6-27B-MTP-GGUF:UD-Q6_K_XL

Files and context on RTX 5090

Each file of Qwen3.6-27B-MTP-GGUF (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)
BF16	54.7 GB	Doesn't fit	—	—
UD-Q8_K_XL	35.8 GB	Doesn't fit	—	—
Q8_0	29.0 GB	Doesn't fit	—	—
UD-Q6_K_XL	26.0 GB	Tight	39K	78K
Q6_K	22.9 GB	Fits	86K	172K
UD-Q5_K_XL	20.4 GB	Fits	124K	247K
Q5_K_M	19.8 GB	Fits	131K	256K
Q5_K_S	19.3 GB	Fits	140K	256K
UD-Q4_K_XL	17.9 GB	Fits	160K	256K
Q4_1	17.5 GB	Fits	165K	256K
Q4_K_M	17.1 GB	Fits	172K	256K
IQ4_NL	16.3 GB	Fits	183K	256K
Q4_K_S	16.1 GB	Fits	187K	256K
Q4_0	16.1 GB	Fits	188K	256K
IQ4_XS	15.7 GB	Fits	193K	256K
UD-Q3_K_XL	14.8 GB	Fits	207K	256K
Q3_K_M	13.8 GB	Fits	221K	256K
Q3_K_S	12.6 GB	Fits	239K	256K
UD-IQ3_XXS	12.2 GB	Fits	245K	256K
UD-Q2_K_XL	12.0 GB	Fits	247K	256K
UD-IQ2_M	11.0 GB	Fits	256K	256K
UD-IQ2_XXS	9.57 GB	Fits	256K	256K

Qwen3.6-27B-MTP-GGUF 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.