← Back to community benchmarks
gemma-4-E2B-it
Performance
32k
tokens
1,248
PP tok/s
39.3
TG tok/s
26254
TTFT (ms)
5.2
Peak mem (GB)
Hardware
Chip
M1 Pro (14c)
Memory
16 GB
GPU Cores
14
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,487 | 53.8 | 4.7 GB | view |
| 4k | 1,505 | 51.9 | 4.7 GB | view |
| 8k | 1,478 | 48.8 | 4.8 GB | view |
| 16k | 1,397 | 44.9 | 4.8 GB | view |
| 32k | 1,248 | 39.3 | 5.2 GB | current |
| 64k | 1,025 | 30.5 | 5.9 GB | view |
| 128k | 726.9 | 21.6 | 7.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 53.8 | 1.00× |
| 2× | 85.0 | 1.58× |
| 4× | 98.2 | 1.83× |