← Back to community benchmarks
gemma-4-E4B-it
Performance
1k
tokens
522.5
PP tok/s
20.0
TG tok/s
1960
TTFT (ms)
6.0
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
16 GB
GPU Cores
10
Software
oMLX
v0.3.6
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 522.5 | 20.0 | 6.0 GB | current |
| 4k | 612.8 | 21.3 | 6.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.0 | 1.00× |
| 2× | 42.9 | 2.15× |
| 4× | 64.4 | 3.22× |