← Back to community benchmarks
gemma-4-e4b-it
Performance
128k
tokens
616.3
PP tok/s
15.7
TG tok/s
212685
TTFT (ms)
13.7
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
64 GB
GPU Cores
16
Software
oMLX
v0.3.9rc1
macOS
macOS 26.5
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,021 | 71.9 | 9.1 GB | view |
| 4k | 1,022 | 66.3 | 9.3 GB | view |
| 8k | 1,005 | 60.1 | 9.4 GB | view |
| 16k | 965.0 | 50.6 | 9.7 GB | view |
| 32k | 880.8 | 38.5 | 10.4 GB | view |
| 64k | 768.0 | 24.1 | 11.3 GB | view |
| 128k | 616.3 | 15.7 | 13.7 GB | current |
| 195k | 500.9 | 11.1 | 16.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 71.9 | 1.00× |
| 2× | 45.1 | 0.63× |
| 4× | 77.6 | 1.08× |
| 8× | 89.9 | 1.25× |