← Back to community benchmarks
gemma-4-e4b
Performance
128k
tokens
715.9
PP tok/s
24.5
TG tok/s
183081
TTFT (ms)
9.9
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.9rc1
macOS
macOS 26.4.1
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,042 | 64.8 | 5.6 GB | view |
| 4k | 1,236 | 63.7 | 5.7 GB | view |
| 8k | 1,228 | 61.6 | 5.8 GB | view |
| 16k | 1,119 | 56.4 | 6.0 GB | view |
| 32k | 962.2 | 48.2 | 6.7 GB | view |
| 64k | 865.3 | 36.6 | 7.6 GB | view |
| 128k | 715.9 | 24.5 | 9.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 64.8 | 1.00× |
| 2× | 124.5 | 1.92× |
| 4× | 150.7 | 2.33× |
| 8× | 165.5 | 2.55× |