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.

Model

Context length

Nex N2.6 (321B-A18B) at 8K context

Quantization	Weights	KV cache + overhead	Total	Smallest hardware that holds it
`Q2_K` Low	129 GB	0.7 GB	129 GB	Apple M2 Ultra (192 GB) Apple M3 Ultra (256 GB) Apple M3 Ultra (512 GB) 25 in all
`Q3_K_M` Low	157 GB	0.7 GB	158 GB	Apple M3 Ultra (256 GB) Apple M3 Ultra (512 GB) 20 in all
`Q4_K_M` Good	196 GB	0.7 GB	197 GB	Apple M3 Ultra (512 GB) 12 in all
`Q5_K_M` High	228 GB	0.7 GB	229 GB	Apple M3 Ultra (512 GB) 9 in all
`Q6_K` High	263 GB	0.7 GB	264 GB	Apple M3 Ultra (512 GB) 7 in all
`Q8_0` Very high	341 GB	0.7 GB	341 GB	Apple M3 Ultra (512 GB) 6 in all
`F16` Full	643 GB	0.7 GB	643 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.