← Back to community benchmarks
gemma-4-31B-it-OptiQ
Performance
64k
tokens
206.3
PP tok/s
6.7
TG tok/s
317625
TTFT (ms)
35.6
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 647.7 | 21.4 | 22.3 GB | view |
| 4k | 592.0 | 19.8 | 25.0 GB | view |
| 8k | 544.5 | 18.0 | 25.5 GB | view |
| 16k | 496.7 | 17.2 | 26.6 GB | view |
| 32k | 424.1 | 11.2 | 29.6 GB | view |
| 64k | 206.3 | 6.7 | 35.6 GB | current |
| 128k | 164.1 | 5.3 | 47.6 GB | view |
| 195k | 126.0 | 4.1 | 62.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.4 | 1.00× |
| 2× | 16.2 | 0.76× |
| 4× | 21.2 | 0.99× |
| 8× | 24.9 | 1.16× |