codingagents · sizes: 30B-A3B (MoE)
| machine | best fit | verdict | speed |
|---|---|---|---|
| Apple M1 · 8GB | 30B-A3B (MoE) Q4_K_M | ❌ Won't fit | — |
| Apple M1 Pro · 16GB | 30B-A3B (MoE) Q4_K_M | ❌ Won't fit | — |
| Apple M2 · 16GB | 30B-A3B (MoE) Q4_K_M | ❌ Won't fit | — |
| Apple M3 Pro · 18GB | 30B-A3B (MoE) Q4_K_M | ❌ Won't fit | — |
| Apple M2 Pro · 32GB | 30B-A3B (MoE) Q6_K | 🐢 CPU-only (slow) | 5–8 tok/s |
| Apple M4 Pro · 48GB | 30B-A3B (MoE) Q8_0 | ✅ Runs (tight) | 40–65 tok/s |
| Apple M3 Max · 36GB | 30B-A3B (MoE) Q4_K_M | ✅ Runs (tight) | 85–140 tok/s |
| Apple M2 Max · 64GB | 30B-A3B (MoE) Q8_0 | ✅ Runs comfortably | 55–90 tok/s |
| Apple M4 Max · 128GB | 30B-A3B (MoE) Q8_0 | ✅ Runs comfortably | 65–110 tok/s |
| Apple M2 Ultra · 192GB | 30B-A3B (MoE) Q8_0 | ✅ Runs comfortably | 90–150 tok/s |
| NVIDIA GeForce RTX 3060 · 12GB VRAM | 30B-A3B (MoE) Q6_K | 🐢 CPU-only (slow) | — |
| NVIDIA GeForce RTX 4070 · 12GB VRAM | 30B-A3B (MoE) Q6_K | 🐢 CPU-only (slow) | — |
| NVIDIA GeForce RTX 4090 · 24GB VRAM | 30B-A3B (MoE) Q4_K_M | ✅ Runs (tight) | 200–330 tok/s |
| NVIDIA GeForce RTX 5080 · 16GB VRAM | 30B-A3B (MoE) Q4_K_M | ⚠️ Partial GPU offload | 60–100 tok/s |
| NVIDIA GeForce RTX 5090 · 32GB VRAM | 30B-A3B (MoE) Q6_K | ✅ Runs (tight) | 270–440 tok/s |
| AMD Radeon RX 7900 XTX · 24GB VRAM | 30B-A3B (MoE) Q4_K_M | ✅ Runs (tight) | 180–300 tok/s |
Shopping for a machine? What hardware do I need to run Qwen3 Coder 30B-A3B? →
Check your exact rig: open the checker →
8 models share Qwen3 Coder 30B-A3B's KV-cache geometry — they size memory the same way, so a rig that fits one tends to fit its same-size kin.
Explore the whole catalog by architecture: open the graph →