← Back to community benchmarks
gemma-4-31b
Performance
8k
tokens
93.6
PP tok/s
6.1
TG tok/s
87514
TTFT (ms)
33.9
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.4
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 99.5 | 7.0 | 32.1 GB | view |
| 4k | 100.3 | 6.6 | 33.6 GB | view |
| 8k | 93.6 | 6.1 | 33.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 7.0 | 1.00× |
| 2× | 13.4 | 1.91× |
| 4× | 18.0 | 2.57× |
| 8× | 18.6 | 2.66× |