← Back to community benchmarks
gemma-4-e4b-it
Performance
64k
tokens
795.0
PP tok/s
37.1
TG tok/s
82440
TTFT (ms)
7.1
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
24 GB
GPU Cores
16
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,024 | 67.6 | 5.6 GB | view |
| 4k | 1,019 | 64.9 | 5.7 GB | view |
| 8k | 1,018 | 62.2 | 5.8 GB | view |
| 16k | 981.3 | 56.8 | 6.0 GB | view |
| 32k | 909.8 | 48.2 | 6.7 GB | view |
| 64k | 795.0 | 37.1 | 7.1 GB | current |
| 128k | 634.0 | 24.8 | 8.8 GB | view |
| 195k | 525.8 | 18.9 | 10.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 67.6 | 1.00× |
| 2× | 111.1 | 1.64× |
| 4× | 127.9 | 1.89× |
| 8× | 139.0 | 2.06× |