What is the minimum hardware to run Mixtral 8x7B?

Nvidia GeForce RTX 5090 (32GB) is the lightest device that runs Mixtral 8x7B, using ~28.9 GB at Q4_K_M of its ~31 GB usable. Anything with less usable memory cannot load it.

How much memory does Mixtral 8x7B need?

At Q4_K_M it needs about 28.9 GB (weights ~26.49 GB + KV cache + overhead at 4k context). At Q8_0 budget ~48.6 GB.

Can I run Mixtral 8x7B on a Mac?

Yes. Mixtral 8x7B runs on Apple Silicon with enough unified memory; the ranked list below shows the lightest Mac that fits.

Can I run Mixtral 8x7B on a phone?

No. Mixtral 8x7B is too large for current phone memory at Q4_K_M; stick to a laptop, desktop GPU, or Mac.

Hardware guide · Mistral

What hardware do you need to run Mixtral 8x7B?

Mixtral 8x7B needs about 28.9 GB to run at Q4_K_M, so the lightest hardware that runs it is Nvidia GeForce RTX 5090 (32GB). 8 of 40 devices tested can run it.

Needs (Q4_K_M): ~28.9 GB
Devices that run it: 8
Too small: 32
Lightest that runs it: ~31 GB

Recommended Q4_K_M

Apple M4 Max (64GB) Yes

Runs at Q4_K_M using ~28.9 GB of ~48 GB usable.

What kind of hardware runs Mixtral 8x7B

Of the 8 devices that run it, here is the split by hardware class.

2 discrete GPUs 6 Macs

Hardware that runs Mixtral 8x7B

Ranked by usable memory, lightest first. Prices are approximate street prices for the device itself (a GPU is the card alone; a Mac is the whole machine). Tok/s is a bandwidth estimate; see methodology.

Compatible devices 8 total

Too small for Mixtral 8x7B

See the full Mixtral 8x7B memory breakdown, or compare it against other models.

Sources

Memory at Q4_K_M (28.9 GB = 26.49 GB weights + KV cache + overhead), validated 2026-08-03. See methodology.