← Back to community benchmarks
gemma-4-12B-it
Performance
64k
tokens
167.5
PP tok/s
13.7
TG tok/s
391376
TTFT (ms)
16.1
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
64 GB
GPU Cores
16
Software
oMLX
v0.4.2.dev1
macOS
macOS 26.5.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 212.0 | 16.8 | 12.8 GB | view |
| 4k | 212.7 | 16.1 | 13.5 GB | view |
| 8k | 197.5 | 14.7 | 13.8 GB | view |
| 16k | 197.9 | 16.2 | 13.9 GB | view |
| 32k | 190.0 | 15.1 | 14.6 GB | view |
| 64k | 167.5 | 13.7 | 16.1 GB | current |
| 128k | 134.7 | 12.3 | 19.3 GB | view |
| 195k | 109.1 | 10.9 | 22.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 16.8 | 1.00× |
| 2× | 28.8 | 1.71× |
| 4× | 34.8 | 2.07× |
| 8× | 37.7 | 2.24× |