2benchesM5 Max
A19 · M5 · 2025 CPU family · 3 nm · First available 2026-03-11
Every configuration
| Configuration | CPU core layout | Bandwidth | GB7 CPU | GB7 Metal | Prefill Q4_0 tok/s | Decode Q4_0 tok/s |
|---|---|---|---|---|---|---|
| 18 CPU / 32 GPU (binned) | 6S + 12P | 460 GB/s | 34,830 | 193,863 | — | — |
| 18 CPU / 40 GPU (full chip) | 6S + 12P | 614 GB/s | 34,830 | 239,008 | 3,220 | 119.9 |
All benchmarks
| Benchmark | Median |
|---|---|
| Geekbench 7 · Multi-core | 34,830 (34,547–35,113, 4 sources) |
| Geekbench 7 · Metal GPU | 216,655 (193,345–239,086, 4 sources) |
| Geekbench AI · GPU · Single | — |
| Geekbench AI · GPU · Half | — |
| Geekbench AI · GPU · Quantized | — |
| Geekbench AI · Neural Engine · Single | — |
| Geekbench AI · Neural Engine · Half | — |
| Geekbench AI · Neural Engine · Quantized | — |
| Prefill 512 · F16 | 3,158 (3,158–3,158, 1 sources) |
| Decode 128 · F16 | 37.1 (37.1–37.1, 1 sources) |
| Prefill 512 · Q8_0 | 3,144 (3,144–3,144, 1 sources) |
| Decode 128 · Q8_0 | 72.4 (72.4–72.4, 1 sources) |
| Prefill 512 · Q4_0 | 3,220 (3,220–3,220, 1 sources) |
| Decode 128 · Q4_0 | 119.9 (119.9–119.9, 1 sources) |
Similar performance
Nearest measured configurations by Geekbench 7 CPU and Metal. Ratios above 1 are faster.
18 CPU / 32 GPU (binned)
- M3 Ultra 28 CPU / 60 GPU (binned), 2025-03: CPU 1.01×, GPU 1.16×
- M4 Max 16 CPU / 40 GPU (full chip), 2024-11: CPU 0.83×, GPU 1.05×
- M2 Ultra 24 CPU / 76 GPU (full chip), 2023-06: CPU 0.81×, GPU 1.15×
- M3 Ultra 32 CPU / 80 GPU (full chip), 2025-03: CPU 1.12×, GPU 1.29×
- M4 Max 14 CPU / 32 GPU (binned), 2024-11: CPU 0.73×, GPU 0.87×
- M5 Pro 18 CPU / 20 GPU (full chip), 2026-03: CPU 0.96×, GPU 0.7×
18 CPU / 40 GPU (full chip)
- M3 Ultra 28 CPU / 60 GPU (binned), 2025-03: CPU 1.01×, GPU 0.94×
- M3 Ultra 32 CPU / 80 GPU (full chip), 2025-03: CPU 1.12×, GPU 1.04×
- M2 Ultra 24 CPU / 76 GPU (full chip), 2023-06: CPU 0.81×, GPU 0.93×
- M4 Max 16 CPU / 40 GPU (full chip), 2024-11: CPU 0.83×, GPU 0.85×
- M4 Max 14 CPU / 32 GPU (binned), 2024-11: CPU 0.73×, GPU 0.71×
- M3 Max 16 CPU / 40 GPU (full chip), 2023-11: CPU 0.7×, GPU 0.68×