← Back to community benchmarks
gemma-4-e4b-it
Performance
32k
tokens
632.1
PP tok/s
25.1
TG tok/s
51839
TTFT (ms)
10.2
Peak mem (GB)
Hardware
Chip
M1 Max (24c)
Memory
32 GB
GPU Cores
24
Software
oMLX
v0.3.4
macOS
macOS 26.3
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 550.4 | 54.3 | 9.3 GB | view |
| 4k | 650.5 | 50.8 | 9.4 GB | view |
| 8k | 663.7 | 45.8 | 9.4 GB | view |
| 16k | 658.2 | 36.9 | 9.6 GB | view |
| 32k | 632.1 | 25.1 | 10.2 GB | current |
| 64k | 577.4 | 15.3 | 11.3 GB | view |
| 128k | 478.5 | 8.9 | 12.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 54.3 | 1.00× |
| 2× | 66.1 | 1.22× |
| 4× | 79.4 | 1.46× |