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

Can I run Qwen3-Coder-Next on an RTX PRO 6000 Blackwell?

Yes, with a GGUF version

The published weights are too big, but the Q8_0 file from Qwen/Qwen3-Coder-Next-GGUF (84.8 GB) fits, with room for ≈ 60K tokens of context.

Full analysis of Qwen3-Coder-Next Everything that runs on RTX PRO 6000 Blackwell Get the Q8_0 file on Hugging Face
llama-server -hf Qwen/Qwen3-Coder-Next-GGUF:Q8_0

Files and context on RTX PRO 6000 Blackwell

Each file of Qwen3-Coder-Next (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 PRO 6000 Blackwell Context (16-bit)Context (8-bit)
As published159 GBDoesn't fit——
F16Qwen/Qwen3-Coder-Next-GGUF159 GBDoesn't fit——
Q8_0Qwen/Qwen3-Coder-Next-GGUF84.8 GBTight60K120K
Q6_KQwen/Qwen3-Coder-Next-GGUF65.5 GBFits256K256K
Q5_K_MQwen/Qwen3-Coder-Next-GGUF56.7 GBFits256K256K
Q5_0Qwen/Qwen3-Coder-Next-GGUF55.0 GBFits256K256K
Q4_K_MQwen/Qwen3-Coder-Next-GGUF48.4 GBFits256K256K

Qwen3-Coder-Next on other hardware

Popular models that run on RTX PRO 6000 Blackwell

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.