What is the minimum hardware to run Qwen3.8 27B?

Apple M5 (32GB) is the lightest device that runs Qwen3.8 27B, using ~17.3 GB at Q4_K_M of its ~21 GB usable. Anything with less usable memory cannot load it.

How much memory does Qwen3.8 27B need?

At Q4_K_M it needs about 17.3 GB (weights ~15.33 GB + KV cache + overhead at 4k context). At Q8_0 budget ~29.1 GB.

Can I run Qwen3.8 27B on a Mac?

Yes. Qwen3.8 27B runs on Apple Silicon with enough unified memory; the ranked list below shows the lightest Mac that fits.

Can I run Qwen3.8 27B on a phone?

No. Qwen3.8 27B is too large for current phone memory at Q4_K_M; stick to a laptop, desktop GPU, or Mac.

Hardware guide · Qwen3.8

What hardware do you need to run Qwen3.8 27B?

Qwen3.8 27B needs about 17.3 GB to run at Q4_K_M, so the lightest hardware that runs it is Apple M5 (32GB). 13 of 43 devices tested can run it.

Needs (Q4_K_M): ~17.3 GB
Devices that run it: 13
Too small: 30
Lightest that runs it: ~21 GB

Recommended Q4_K_M

Apple M5 (32GB) Yes

Runs at Q4_K_M using ~17.3 GB of ~21 GB usable.

What kind of hardware runs Qwen3.8 27B

Of the 13 devices that run it, here is the split by hardware class.

5 discrete GPUs 7 Macs 1 laptop

Hardware that runs Qwen3.8 27B

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 13 total

Apple M5 (32GB) from $1,299

~21 GB usable · uses ~17.3 GB · ~8 tok/s est.

Yes
Nvidia GeForce RTX 4090 (24GB) from $1,599

~23 GB usable · uses ~17.3 GB · ~43 tok/s est.

Yes
Nvidia GeForce RTX 3090 (24GB) from $1,499

~23 GB usable · uses ~17.3 GB · ~40 tok/s est.

Yes
AMD Radeon RX 7900 XTX (24GB) from $999

~23 GB usable · uses ~17.3 GB · ~41 tok/s est.

Yes
32GB RAM Laptop (CPU/iGPU only) from $1,100

~28 GB usable · uses ~17.3 GB · ~3 tok/s est.

Yes
Nvidia GeForce RTX 5090 (32GB) from $1,999

~31 GB usable · uses ~17.3 GB · ~76 tok/s est.

Yes
Apple M4 Pro (48GB) from $2,399

~32 GB usable · uses ~17.3 GB · ~14 tok/s est.

Yes
Apple M5 Pro (48GB) from $2,199

~32 GB usable · uses ~17.3 GB · ~16 tok/s est.

Yes
Apple M4 Max (64GB) from $3,499

~48 GB usable · uses ~17.3 GB · ~28 tok/s est.

Yes
Apple M4 Max (128GB) from $3,499

~96 GB usable · uses ~17.3 GB · ~28 tok/s est.

Yes
AMD Ryzen AI Halo (128GB) from $3,999

~96 GB usable · uses ~17.3 GB · ~11 tok/s est.

Yes
Apple M5 Max (128GB) from $3,599

~96 GB usable · uses ~17.3 GB · ~32 tok/s est.

Yes
Apple M3 Ultra (256GB) from $3,999

~192 GB usable · uses ~17.3 GB · ~43 tok/s est.

Yes

See the full Qwen3.8 27B memory breakdown, or compare it against other models.

Sources

huggingface.co/Qwen
huggingface.co/unsloth
ollama.com

Memory at Q4_K_M (17.3 GB = 15.33 GB weights + KV cache + overhead), validated 2026-09-07. See methodology.

What hardware do you need to run Qwen3.8 27B?

What kind of hardware runs Qwen3.8 27B

Hardware that runs Qwen3.8 27B

Too small for Qwen3.8 27B

Sources