Qwen3.8 27B

Qwen3.8 27B loads in 16.5 GB at Q4_K_M and is comfortable from 32 GB of memory, Macs included from the MacBook Air M5 24GB. Below: the full memory math, the cheapest card that runs it well, and the fastest machine we track.

PARAMETERS
27B
FORMAT
Q4_K_M
MIN MEMORY
24 GB
BEST FOR
Coding, Agent, Vision, Long context

Memory math

Weights (Q4_K_M)
16.5 GB
+ KV cache (16k)
~1.0 GB
Total at 16k
~17.5 GB
Comfortable from
32 GB

KV = fp16 estimate (q8_0 cache roughly halves it). "Comfortable" = weights + KV within the engine's tiered budget (~70-85% of memory).

Hardware snapshot

Cheapest GPU
~26 tok/s est. — ~$550 used
Fastest
~52 tok/s est.
Runs on a Mac?
~7 tok/s est.

Go deeper

More Qwen models

Frequently asked questions

How much memory does Qwen3.8 27B need?

16.5 GB for the Q4_K_M weights, plus ~1.0 GB of KV-cache at 16k context — about 17.5 GB total. Comfortable from 32 GB of VRAM or unified memory.

Does Qwen3.8 27B run on a Mac?

Yes — from the MacBook Air M5 24GB (~7 tok/s est.). Unified memory means the RAM budget is the only limit.

What is the cheapest GPU for Qwen3.8 27B?

The AMD Radeon RX 7900 XT is the cheapest tracked card that runs Qwen3.8 27B comfortably — ~26 tok/s est. at ~$550 used.

Cite this page

ModelFit: Qwen3.8 27B — specs, memory math and hardware verdicts.
https://modelfit.io/models/qwen3.8-27b/ (dataset updated 2026-09-03, CC BY 4.0).