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

Can I run Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP-GGUF on an RTX 5090?

Yes

Its BF16 file (18.4 GB) fits in the 28.8 GB an RTX 5090 can use, with room for ≈ 256K tokens of context.

Full analysis of Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP-GGUF Everything that runs on RTX 5090 Get the BF16 file on Hugging Face
ollama run hf.co/DavidAU/Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP-GGUF:BF16

Files and context on RTX 5090

Each file of Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-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.

FileSizeOn RTX 5090 Context (16-bit)Context (8-bit)
BF1618.4 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-plusIQ-TOOLS-NEO-MAX-MTP-Q8_010.7 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-plusIQ-NEO-MAX-MTP-Q8_010.7 GBFits256K256K
Q8_010.7 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-Q8_010.5 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-plusIQ-TOOLS-NEO-MAX-MTP-Q6_K8.76 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-plusIQ-NEO-MAX-MTP-Q6_K8.76 GBFits256K256K
Q6_K8.76 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-Q6_K8.56 GBFits256K256K
Q5_K_M7.84 GBFits256K256K
Q5_K_S7.67 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-Q5_K_M7.67 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-Q5_K_S7.51 GBFits256K256K
Q4_K_M6.98 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-Q4_K_M6.83 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-MTP-IQ4_NL6.76 GBFits256K256K
Q4_K_S6.69 GBFits256K256K
IQ4_NL6.62 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-Q4_K_S6.55 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-MTP-IQ4_XS6.53 GBFits256K256K
IQ4_XS6.40 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-MTP-IQ3_M5.72 GBFits256K256K
IQ3_M5.62 GBFits256K256K
Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-MTP-IQ2_M4.94 GBFits256K256K
IQ2_M4.94 GBFits256K256K

Qwen3.5-9B-The-Defiant-Fable-Uncensored-Heretic-NEO-IMATRIX-MAX-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.