← Back to community benchmarks
gemma-4-31b-it
Performance
4k
tokens
162.2
PP tok/s
16.9
TG tok/s
25257
TTFT (ms)
19.4
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
48 GB
GPU Cores
40
Software
oMLX
v0.3.4
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 171.4 | 18.7 | 17.6 GB | view |
| 4k | 162.2 | 16.9 | 19.4 GB | current |
| 8k | 163.2 | 13.6 | 19.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 18.7 | 1.00× |
| 2× | 26.1 | 1.40× |
| 4× | 29.8 | 1.59× |