Can I Run Aya Expanse 32B on Apple M3 Pro?
作者： Jakub Rusinowski · 最后更新： 2026年9月29日
Yes. Aya Expanse 32B needs 21.6 GB at 8K context (19.5 GB weights + 2.1 GB KV cache/overhead) at Q4_K_M and the Apple M3 Pro has 27 GB usable, leaving about 5.4 GB of headroom, at roughly 4 tok/s (estimated) with room for up to 32K of context.
See what else this hardware can run →
The numbers
| Model | Aya Expanse 32B |
| Parameters | 32B |
| Quantization | Q4_K_M |
| VRAM needed (8K context) | 21.6 GB |
| Apple M3 Pro usable memory | 27 GB |
| Verdict | Runs |
| Max context | 32K |
| Estimated speed | ~4 tok/s |
Aya Expanse 32B on the Apple M3 Pro: VRAM by quantization
| Quant | VRAM needed | Fits 36 GB? | Max context | Whole PC |
|---|---|---|---|---|
| F16 | 66.7 GB | ✗ No | — | — |
| Q8_0 | 36.5 GB | ✗ No | — | Build for this |
| Q6_K | 28.6 GB | ✗ No | — | — |
| Q5_K_M | 25 GB | ✓ Yes | 16K | Build for this |
| Q4_K_M | 21.6 GB | ✓ Yes | 32K | Build for this |
| Q3_K_M | 15.9 GB | ✓ Yes | 64K | — |
| Q2_K | 12.8 GB | ✓ Yes | 64K | — |
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 36 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 Pro: 8K 21.6 GB · 32K 25.7 GB · 125K 41.3 GB ✗. Past 125K it no longer fits 36 GB — a q8_0 KV cache buys roughly half of that back, which the calculator will price for you.
What to do next
Nothing to change — run it at Q4_K_M. If you want more capability, check whether the next size up also fits.
No compatible GPU? Aya Expanse 32B on 32 GB of system RAM, CPU only: runs comfortably.
Running Aya Expanse 32B on this card:
Aya Expanse 32B needs ~22 GB but this GPU has 36 GB. Rent a RTX 4090 (24 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.
Other Aya Expanse sizes on the Apple M3 Pro
- Aya Expanse 8B — runs (6.7 GB)
Aya Expanse 32B on nearby hardware
- Apple M5 (32 GB) — runs
- NVIDIA RTX 6000 Ada Generation (48 GB) — runs
All Aya Expanse sizes on the Apple M3 Pro | Aya Expanse 32B model page | Apple M3 Pro GPU page | VRAM calculator | Check your hardware