Performance
8k
tokens
296.1
PP tok/s
33.6
TG tok/s
27663
TTFT (ms)
4.6
Peak mem (GB)
Hardware
Chip
M4 (8c)
Memory
16 GB
GPU Cores
8
Software
oMLX
v0.2.13
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 309.0 | 38.6 | 3.9 GB | view |
| 4k | 306.6 | 35.9 | 4.2 GB | view |
| 8k | 296.1 | 33.6 | 4.6 GB | current |
| 16k | 260.7 | 25.6 | 4.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 38.6 | 1.00× |
| 2× | 50.8 | 1.32× |
| 4× | 62.5 | 1.62× |
| 8× | 64.0 | 1.66× |