Apple M1 Max (32-core GPU, 64GB) × Phi-4 14B
Can the Apple M1 Max (32-core GPU, 64GB) run Phi-4 14B?
Yes
Yes. The Apple M1 Max (32-core GPU, 64GB) can run Phi-4 14B at up to F16 (needs ~32.8 GB of 48.0 GB usable), at an estimated ~26.2 tokens/sec at Q4_K_M.
Speed is estimated from memory bandwidth calibrated with a measured 61.19 tok/s Llama 2 7B Q4_0 result (source). No direct public benchmark of this exact pair is recorded yet.
By quantization (8K context)
| Quant | Memory needed | Fits? | Est. tok/s |
|---|---|---|---|
| Q3_K_M | 9.9 GB | ✅ | ~32.6 |
| Q4_K_M | 11.7 GB | ✅ | ~26.2 |
| Q5_K_M | 13.3 GB | ✅ | ~22.3 |
| Q6_K | 15.0 GB | ✅ | ~19.3 |
| Q8_0 | 18.6 GB | ✅ | ~15.0 |
| F16 | 32.8 GB | ✅ | ~8.0 |
Fastest hardware for Phi-4 14B
- NVIDIA GeForce RTX 5090 — ~124 tok/s at Q4_K_M
- NVIDIA GeForce RTX 4090 — ~79.9 tok/s at Q4_K_M
- AMD Radeon RX 7900 XTX — ~79.1 tok/s at Q4_K_M
- NVIDIA GeForce RTX 5080 — ~78.1 tok/s at Q4_K_M
- NVIDIA GeForce RTX 5070 Ti — ~75.9 tok/s at Q4_K_M
No suitable GPU? Rent one by the hour on Vast.ai or RunPod (affiliate links).
Get new models & benchmarks in your inbox
No spam. Unsubscribe anytime by replying to any email.