← Back to community benchmarks
gemma-4-31B-it
Performance
32k
tokens
511.3
PP tok/s
12.4
TG tok/s
64091
TTFT (ms)
39.0
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 613.0 | 15.3 | 31.8 GB | view |
| 4k | 659.5 | 14.9 | 34.5 GB | view |
| 8k | 622.1 | 14.6 | 35.0 GB | view |
| 16k | 568.6 | 13.8 | 36.1 GB | view |
| 32k | 511.3 | 12.4 | 39.0 GB | current |
| 64k | 415.1 | 10.3 | 45.1 GB | view |
| 128k | 285.9 | 7.3 | 57.1 GB | view |
| 195k | 203.3 | 5.7 | 72.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 15.3 | 1.00× |
| 2× | 22.5 | 1.47× |
| 4× | 29.2 | 1.91× |
| 8× | 34.7 | 2.27× |