← Back to community benchmarks
gemma-4-E4B-it-oQ8
Performance
16k
tokens
2,101
PP tok/s
62.5
TG tok/s
7800
TTFT (ms)
9.5
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,533 | 73.5 | 8.9 GB | view |
| 4k | 2,513 | 70.4 | 9.1 GB | view |
| 8k | 2,271 | 67.7 | 9.2 GB | view |
| 16k | 2,101 | 62.5 | 9.5 GB | current |
| 32k | 1,782 | 54.5 | 10.2 GB | view |
| 64k | 1,496 | 42.1 | 11.0 GB | view |
| 128k | 1,185 | 24.8 | 13.3 GB | view |
| 195k | 934.1 | 22.4 | 16.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 73.5 | 1.00× |
| 2× | 101.0 | 1.37× |
| 4× | 162.0 | 2.20× |
| 8× | 181.2 | 2.47× |