Can I Run Qwen3-Coder on Apple M3 Ultra?
作者： Jakub Rusinowski · 最后更新： 2026年9月29日
Yes, comfortably — Qwen3-Coder 480B-A35B (MoE) at Q4_K_M needs 291.6 GB at 8K context (289.8 GB weights + 1.8 GB KV cache/overhead), leaving ~92.4 GB of the Apple M3 Ultra's 384 GB usable memory, at ~20 tok/s (est.), with room for up to 256K context.
See what else this hardware can run →
作为亚马逊联盟成员，我们从符合条件的购买中获得收入。云 GPU 链接为推荐链接——我们可能获得佣金，您无需额外付费。
Apple M3 Ultra Specs
| VRAM | 512 GB unified memory |
| Memory Bandwidth | 819 GB/s |
Qwen3-Coder 480B-A35B (MoE) on the Apple M3 Ultra: VRAM by quantization
| Quant | VRAM needed | Fits 512 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 961.8 GB | ✗ No | — | — |
| Q8_0 | 511.8 GB | ✗ No | — | — |
| Q6_K | 395.4 GB | ✗ No | — | — |
| Q5_K_M | 342 GB | ✓ Yes | 256K | — |
| Q4_K_M | 291.6 GB | ✓ Yes | 256K | — |
| Q3_K_M | 206.4 GB | ✓ Yes | 256K | — |
| Q2_K | 159.6 GB | ✓ Yes | 256K | — |
Assumes an 8K-token context with an f16 KV cache. A longer window needs more; a quantized KV cache needs less. “Max context” is the largest window that still fits in 512 GB of unified memory. Figures are estimates from parameter count, quantization and memory bandwidth — the analyzer lets you tune KV-cache quant and context.
Context costs VRAM too. At Q4_K_M on the Apple M3 Ultra: 8K 291.6 GB · 32K 294.8 GB · 128K 307.2 GB. Every window this model publishes still fits 512 GB.
No compatible GPU? Qwen3-Coder on 16 GB of system RAM, CPU only: runs, but it is tight.
Qwen3-Coder Sizes That Fit the Apple M3 Ultra (at 8K context)
| Qwen3-Coder 480B-A35B (MoE) | Q4_K_M · 291.6 GB at 8K context · ~20 tok/s (est.) |
| Qwen3-Coder-Next (80B-A3B MoE) | Q4_K_M · 51.6 GB at 8K context · ~102 tok/s (est.) |
| Qwen3-Coder 30B-A3B (MoE) | Q4_K_M · 20 GB at 8K context · ~120 tok/s (est.) |
Qwen3-Coder needs ~292 GB but this GPU has 512 GB. Rent a 8× H100 (640 GB)-class GPU by the hour instead of buying one:
Affiliate links — we may earn a commission if you sign up, at no extra cost to you.
Cloud rates verified 2026-07 — estimates, and marketplace prices vary. Buying price is GPU MSRP only, not a full PC.
FAQ
Is the Apple M3 Ultra enough to run Qwen3-Coder locally?
Yes, comfortably — Qwen3-Coder 480B-A35B (MoE) at Q4_K_M needs 291.6 GB at 8K context (289.8 GB weights + 1.8 GB KV cache/overhead), leaving ~92.4 GB of the Apple M3 Ultra's 384 GB usable memory, at ~20 tok/s (est.), with room for up to 256K context.
What's the best Qwen3-Coder size for the Apple M3 Ultra?
Qwen3-Coder 480B-A35B (MoE) at Q4_K_M quantization needs 291.6 GB at 8K context (289.8 GB weights + 1.8 GB KV cache/overhead), estimated ~20 tokens/sec, up to 256K context.
Every Qwen3-Coder size on the Apple M3 Ultra
| Size | VRAM at 8K | Verdict | Speed |
|---|---|---|---|
| Qwen3-Coder 480B-A35B (MoE) | 291.6 GB | Runs | ~20 tok/s |
| Qwen3-Coder-Next (80B-A3B MoE) | 51.6 GB | Runs | ~102 tok/s |
Popular Models on the Apple M3 Ultra
VRAM Tier
Troubleshooting
Buying Guide
← Can I Run It? | Qwen3-Coder Model Page | Apple M3 Ultra GPU Page | VRAM calculator | Check Your Hardware