← Back to community benchmarks
gemma-4-31b
Performance
8k
tokens
287.4
PP tok/s
6.3
TG tok/s
28511
TTFT (ms)
34.0
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 290.4 | 7.8 | 32.1 GB | view |
| 4k | 309.0 | 7.3 | 33.6 GB | view |
| 8k | 287.4 | 6.3 | 34.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 7.8 | 1.00× |
| 2× | 13.4 | 1.72× |
| 4× | 17.8 | 2.28× |
| 8× | 19.0 | 2.44× |