← Back to community benchmarks
gemma-4-E4B-it
Performance
32k
tokens
1,012
PP tok/s
38.1
TG tok/s
32372
TTFT (ms)
7.3
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
64 GB
GPU Cores
38
Software
oMLX
v0.5.1
macOS
macOS 26.5.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,034 | 55.7 | 6.1 GB | view |
| 4k | 1,093 | 53.9 | 6.2 GB | view |
| 8k | 1,134 | 50.9 | 6.4 GB | view |
| 16k | 1,094 | 43.8 | 6.6 GB | view |
| 32k | 1,012 | 38.1 | 7.3 GB | current |
| 64k | 930.8 | 29.7 | 8.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 55.7 | 1.00× |
| 2× | 100.6 | 1.81× |
| 4× | 137.2 | 2.46× |
| 8× | 121.1 | 2.17× |