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

Can I run Qwen3.5-9B on an RTX 4060?

Yes, with a GGUF version

The published weights are too big, but the UD-Q5_K_XL file from unsloth/Qwen3.5-9B-GGUF (6.74 GB) fits, with room for ≈ 12K tokens of context.

Full analysis of Qwen3.5-9B Everything that runs on RTX 4060 Get the UD-Q5_K_XL file on Hugging Face
ollama run hf.co/unsloth/Qwen3.5-9B-GGUF:UD-Q5_K_XL

Files and context on RTX 4060

Each file of Qwen3.5-9B (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 4060 Context (16-bit)Context (8-bit)
As published19.3 GBDoesn't fit——
BF16unsloth/Qwen3.5-9B-GGUF17.9 GBDoesn't fit——
UD-Q8_K_XLunsloth/Qwen3.5-9B-GGUF13.0 GBDoesn't fit——
Q8_0unsloth/Qwen3.5-9B-GGUF9.53 GBDoesn't fit——
UD-Q6_K_XLunsloth/Qwen3.5-9B-GGUF8.76 GBDoesn't fit——
Q6_Kunsloth/Qwen3.5-9B-GGUF7.46 GBDoesn't fit——
UD-Q5_K_XLunsloth/Qwen3.5-9B-GGUF6.74 GBTight12K24K
Q5_K_Munsloth/Qwen3.5-9B-GGUF6.58 GBTight17K34K
Q5_K_Sunsloth/Qwen3.5-9B-GGUF6.36 GBTight23K47K
UD-Q4_K_XLunsloth/Qwen3.5-9B-GGUF5.97 GBTight35K70K
Q4_1unsloth/Qwen3.5-9B-GGUF5.84 GBTight39K78K
Q4_K_Munsloth/Qwen3.5-9B-GGUF5.68 GBFits44K87K
Q4_K_Sunsloth/Qwen3.5-9B-GGUF5.39 GBFits52K105K
Q4_0unsloth/Qwen3.5-9B-GGUF5.38 GBFits53K105K
IQ4_NLunsloth/Qwen3.5-9B-GGUF5.37 GBFits53K106K
IQ4_XSunsloth/Qwen3.5-9B-GGUF5.17 GBFits59K118K
UD-Q3_K_XLunsloth/Qwen3.5-9B-GGUF5.05 GBFits62K125K
Q3_K_Munsloth/Qwen3.5-9B-GGUF4.67 GBFits74K148K
Q3_K_Sunsloth/Qwen3.5-9B-GGUF4.32 GBFits84K169K
UD-Q2_K_XLunsloth/Qwen3.5-9B-GGUF4.12 GBFits90K180K
UD-IQ3_XXSunsloth/Qwen3.5-9B-GGUF4.02 GBFits93K187K
UD-IQ2_Munsloth/Qwen3.5-9B-GGUF3.65 GBFits104K209K
UD-IQ2_XXSunsloth/Qwen3.5-9B-GGUF3.19 GBFits118K236K

Qwen3.5-9B on other hardware

Popular models that run on RTX 4060

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.