← Back to community benchmarks
gemma-4-E4B-it-oQ8
Performance
64k
tokens
1,195
PP tok/s
25.6
TG tok/s
54860
TTFT (ms)
11.0
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,505 | 50.3 | 8.9 GB | view |
| 4k | 1,533 | 48.4 | 9.1 GB | view |
| 8k | 1,518 | 46.3 | 9.2 GB | view |
| 16k | 1,469 | 42.4 | 9.5 GB | view |
| 32k | 1,369 | 35.8 | 10.2 GB | view |
| 64k | 1,195 | 25.6 | 11.0 GB | current |
| 128k | 925.3 | 17.2 | 13.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 50.3 | 1.00× |
| 2× | 88.2 | 1.75× |
| 4× | 108.5 | 2.16× |