App	Type	Opens it as	Systems	Engine	Start
vLLM	Server	Original weights	WSLWindowsLinux	vLLM	`vllm serve beipo-ai/NorthCUA-Pro`
SGLang	Server	Original weights	WSLWindowsLinux	SGLang	`python3 -m sglang.launch_server --model-path beipo-ai/NorthCUA-Pro`
Docker Model Runner	CLI + API	Original weights	macOSWindowsLinux	llama.cpp, vLLM	`docker model run hf.co/beipo-ai/NorthCUA-Pro`
Text Generation WebUI	Desktop app	Original weights	macOSWindowsLinux	llama.cpp, Transformers, ExLlamaV3	Get Text Generation WebUI

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory needed

Weights plus KV cache and runtime overhead. Leave a little extra free for the operating system and other apps. Try another context length.

KV cache: hybrid: only the full-attention layers (15 of 60) keep a KV cache; the rest use linear attention.

Quantization	Quality	Weights	Total at 8K context	Total at 32K context	Download
`Q2_K`	Low	159 GB	160 GB	160 GB	All files (opens in a new tab)
`Q3_K_M`	Low	194 GB	195 GB	196 GB	All files (opens in a new tab)
`Q4_K_M`	Good	242 GB	243 GB	244 GB	All files (opens in a new tab)
`Q5_K_M`	High	282 GB	283 GB	283 GB	All files (opens in a new tab)
`Q6_K`	High	325 GB	326 GB	327 GB	All files (opens in a new tab)
`Q8_0`	Very high	421 GB	421 GB	422 GB	All files (opens in a new tab)
`F16`	Full	794 GB	794 GB	795 GB	Original weights (opens in a new tab)

Best hardware for NorthCUA Pro

Quality	Hardware	Quant	Memory	Speed	Runs
75Great quality (A) · 75 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 88/100 (quality, fit, speed and memory use) sets the ranking.	Apple M3 Ultra (512 GB)	`Q5_K_M`	313 GB / 384 GB	36.7 tok/s	Runs great
54Fair quality (C) · 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	Apple M3 Ultra (256 GB)	`Q2_K`	175 GB / 192 GB	66.3 tok/s	Runs welltight fit
75Great quality (A) · 75 of 100: how capable the model is at Q5_K_M, whatever it runs on. Match score 89/100 (quality, fit, speed and memory use) sets the ranking.	Apple M5 Ultra (512 GB)	`Q5_K_M`	313 GB / 384 GB	50.7 tok/s	Runs great
74Great quality (A) · 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.	AMD Instinct MI325X	`Q4_K_M`	256 GB / 256 GB	165 tok/s	Runs welltight fit
74Great quality (A) · 74 of 100: how capable the model is at Q4_K_M, whatever it runs on. Match score 87/100 (quality, fit, speed and memory use) sets the ranking.	AMD Instinct MI350X	`Q4_K_M`	257 GB / 288 GB	186 tok/s	Runs welltight fit
65Good quality (B) · 65 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 84/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip	`Q3_K_M`	208 GB / 252 GB	211 tok/s	Runs great
54Fair quality (C) · 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	AMD Instinct MI300X	`Q2_K`	169 GB / 192 GB	201 tok/s	Runs welltight fit
54Fair quality (C) · 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	Apple M5 Ultra (256 GB)	`Q2_K`	175 GB / 192 GB	88.6 tok/s	Runs welltight fit
54Fair quality (C) · 54 of 100: how capable the model is at Q2_K, whatever it runs on. Match score 78/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA B200	`Q2_K`	169 GB / 180 GB	239 tok/s	Runs welltight fit
65Good quality (B) · 65 of 100: how capable the model is at Q3_K_M, whatever it runs on. Match score 76/100 (quality, fit, speed and memory use) sets the ranking.	AMD EPYC 7002 / 7003 series NAS (256 GB)	`Q3_K_M`	214 GB / 230 GB	10.3 tok/s	Runs welltight fit

Check it on your own hardware

Run it on a cluster

Devices that can’t run NorthCUA Pro alone can run it together. These example clusters do:

Two DGX Sparks38.1 tok/s at Q3_K_M with MLX distributed

Try it with your own devices

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.