Model family · 3 sizes

Qwen3-Coder: which size runs locally?

Qwen3-Coder comes in 3 sizes, from 30.5B to 480B. Some sizes are Mixture-of-Experts, so they run faster than their memory footprint suggests. Here is each size with its Q4_K_M weight, the memory it needs, and the hardware that runs it.

Sizes: 3
Smallest: 30.5B
Largest: 480B
Runs from: 24GB

The Qwen3-Coder lineup

Smallest first Q4_K_M · min memory

"Needs" is the sourced minimum memory for Q4_K_M with a small context. Larger context needs more.

Which Qwen3-Coder fits your memory

8GB

No Qwen3-Coder size fits 8GB; even Qwen3-Coder 30B-A3B needs more.

16GB

No Qwen3-Coder size fits 16GB; even Qwen3-Coder 30B-A3B needs more.

24GB

Largest that fits: Qwen3-Coder 30B-A3B (30.5B), best case on Nvidia GeForce RTX 4090 (24GB).

Yes

32GB

Largest that fits: Qwen3-Coder 30B-A3B (30.5B), best case on Nvidia GeForce RTX 5090 (32GB).

Yes

48GB

Largest that fits: Qwen3-Coder 30B-A3B (30.5B), best case on Apple M5 Pro (48GB).

Yes

64GB

Largest that fits: Qwen3-Coder-Next 80B-A3B (80B), best case on Apple M4 Max (64GB). Comfortable up to Qwen3-Coder 30B-A3B (30.5B).

Tight

128GB

Largest that fits: Qwen3-Coder-Next 80B-A3B (80B), best case on Apple M5 Max (128GB).

Yes

256GB

Largest that fits: Qwen3-Coder-Next 80B-A3B (80B), best case on Apple M3 Ultra (256GB).

Yes

Best case means the most capable device at that size (usually a discrete GPU). A Mac at the same size sits roughly one rung lower; see the per-size breakdown on each memory budget page.

FAQ

Which Qwen3-Coder size should I run locally?

Pick the largest size your memory allows. On 24GB (best case) up to Qwen3-Coder 30B-A3B; On 32GB (best case) up to Qwen3-Coder 30B-A3B; On 48GB (best case) up to Qwen3-Coder 30B-A3B; On 64GB (best case) up to Qwen3-Coder-Next 80B-A3B; On 128GB (best case) up to Qwen3-Coder-Next 80B-A3B; On 256GB (best case) up to Qwen3-Coder-Next 80B-A3B. Smaller sizes run faster and leave headroom for context.

What is the smallest Qwen3-Coder model?

Qwen3-Coder 30B-A3B at 30.5B parameters, about 17.28 GB on disk at Q4_K_M. It is the one to use on phones and 8 GB machines.

What is the largest Qwen3-Coder model and what does it need?

Qwen3-Coder 480B-A35B Instruct at 480B (mixture of experts), about 270.14 GB at Q4_K_M. It needs more than a typical 32 GB desktop; a high-memory Mac or multi-GPU rig.

Understand the numbers

Short guides to the ideas behind Qwen3-Coder's memory and quant figures.

Sources

Memory figures are estimates at Q4_K_M. See methodology.