← Back to community benchmarks
gemma-4-31B-it-oQ8
Performance
64k
tokens
435.4
PP tok/s
2.1
TG tok/s
150502
TTFT (ms)
42.3
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.5
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 597.0 | 14.9 | 31.8 GB | view |
| 4k | 617.5 | 14.5 | 33.8 GB | view |
| 8k | 597.8 | 13.0 | 34.3 GB | view |
| 16k | 561.2 | 10.8 | 35.5 GB | view |
| 32k | 537.9 | 5.3 | 37.2 GB | view |
| 64k | 435.4 | 2.1 | 42.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 14.9 | 1.00× |
| 2× | 25.7 | 1.72× |
| 4× | 36.1 | 2.42× |
| 8× | 70.5 | 4.73× |