Hardware · Memory

How much memory does a model need?

Pick an LLM and a context length. We add up the weights at every quantization, the KV cache that grows with context, and runtime overhead, then show the smallest systems and devices that hold it.

NorthCUA Pro at 8K context

Quantization Weights KV cache + overhead Total Smallest hardware that holds it
Q2_K Low 159 GB 0.9 GB 160 GB 20 in all
Q3_K_M Low 194 GB 0.9 GB 195 GB 12 in all
Q4_K_M Good 242 GB 0.9 GB 243 GB 7 in all
Q5_K_M High 282 GB 0.9 GB 283 GB 6 in all
Q6_K High 325 GB 0.9 GB 326 GB 6 in all
Q8_0 Very high 421 GB 0.9 GB 421 GB 4 in all
F16 Full 794 GB 0.9 GB 794 GB Nothing in the catalog holds this on one machine.

Hardware is listed when its usable memory is at least 8% above the total. Macs count roughly two-thirds to three-quarters of unified memory as usable. Each link opens the full “can it run” breakdown with speed.

The rule of thumb

1 · WeightsParameters × bytes eachA 70B model at 4 bits per weight is about 70 × 0.6 ≈ 42 GB. Lower quantization shrinks it and costs some quality.
2 · KV cacheGrows with contextEvery token you keep in context stores keys and values for each layer. Doubling context doubles this part, so long documents cost real memory.
3 · OverheadLeave headroomThe runtime, the operating system and other apps need memory too. Aim for a total at least 10% under the memory your GPU or device can give a model.

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.

Was this page helpful?