← Back to community benchmarks
gemma-4-E2B-it-4bit
Performance
32k
tokens
4,190
PP tok/s
36.2
TG tok/s
7820
TTFT (ms)
5.0
Peak mem (GB)
Hardware
Chip
M5 (10c)
Memory
24 GB
GPU Cores
10
Software
oMLX
v0.3.2
macOS
macOS 26.4
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,136 | 47.9 | 4.5 GB | view |
| 4k | 4,133 | 45.5 | 4.6 GB | view |
| 16k | 4,999 | 41.6 | 4.7 GB | view |
| 32k | 4,190 | 36.2 | 5.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 47.9 | 1.00× |
| 2× | 46.8 | 0.98× |
| 4× | 48.4 | 1.01× |