← Back to community benchmarks
gemma-4-e4b-it
Performance
16k
tokens
2,521
PP tok/s
47.3
TG tok/s
6498
TTFT (ms)
15.7
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.9rc1
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,547 | 54.3 | 15.3 GB | view |
| 4k | 2,648 | 52.4 | 15.4 GB | view |
| 8k | 2,615 | 51.1 | 15.5 GB | view |
| 16k | 2,521 | 47.3 | 15.7 GB | current |
| 32k | 2,272 | 41.8 | 16.1 GB | view |
| 64k | 1,995 | 32.4 | 17.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 54.3 | 1.00× |
| 2× | 50.8 | 0.94× |
| 4× | 97.6 | 1.80× |
| 8× | 191.5 | 3.53× |