← Back to community benchmarks
gemma-4-E4B-it-oQ8
Performance
64k
tokens
3,653
PP tok/s
83.0
TG tok/s
17940
TTFT (ms)
13.4
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.4.2.dev3
macOS
macOS 26.6
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 4,070 | 121.5 | 11.2 GB | view |
| 8k | 4,815 | 111.3 | 11.5 GB | view |
| 16k | 4,815 | 108.7 | 11.8 GB | view |
| 32k | 4,331 | 98.4 | 12.5 GB | view |
| 64k | 3,653 | 83.0 | 13.4 GB | current |
| 128k | 2,682 | 51.5 | 15.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 121.5 | 1.00× |
| 2× | 69.3 | 0.57× |
| 4× | 152.1 | 1.25× |