What GPU do I need to run beipo-ai/NorthCUA-Pro?
396.8B parameters, published in BF16. View on Hugging Face
Set up NorthCUA-ProNorthCUA-Pro is published by beipo-ai on Hugging Face, with 811 downloads and 32 likes to date. It's a Qwen3_5MoeForConditionalGeneration model built for image-text-to-text, published natively in BF16.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 739.1 GB | 886.9 GB | No capable live offer found | – | – |
| FP8 (quantized) | 369.6 GB | 443.5 GB | RTX PRO 6000 | 5 | $6.88/hr |
| INT4 (quantized) | 184.8 GB | 221.7 GB | RTX A6000 | 5 | $2.75/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
NorthCUA-Pro: KV cache by context length
The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.
Hybrid: only the full-attention layers cache per token; the other layers keep a fixed-size state. Cached per token: 30,720 bytes at 16-bit.
| Context | KV cache, one sequence |
|---|---|
| 4K | 0.12 GB |
| 32K | 0.94 GB |
| 128K | 3.8 GB |
| 256K (model maximum) | 7.5 GB |
Computed from the layer, head and window counts in the model's own configuration (beipo-ai/NorthCUA-Pro), at 16 bits per cached value. Each concurrent sequence needs its own cache, on top of the weights.
NorthCUA-Pro: common questions
Can NorthCUA-Pro run on a single GPU?
Not on a desktop card. At BF16 it needs 886.9 GB of VRAM, more than a single 32 GB desktop card holds. No card currently listed on the marketplace both supports BF16 and has enough VRAM for it, so how many it would take is not something this page can answer today.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
Alternatives at this size
Other models for image-text-to-text within about a third of NorthCUA-Pro's 396.8B parameters, from other model lines.
- GLM-5.3-Flash (321.3B, F8_E4M3)
- MiniMax-M3 (427.0B, BF16)
- Qwen3.5-397B-A17B (403.4B, BF16)
- Inkling-Small (266.0B, BF16)
- step3 (321.0B, BF16)
More on NorthCUA-Pro
Fits on a 288 GB GPU at INT4: every model that fits in 288 GB.
Best vision-language models: how NorthCUA-Pro ranks against the rest.
Related reading: H100 pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.
