← Back to community benchmarks
gemma-4-31B-it-oQ8e
Performance
1k
tokens
188.5
PP tok/s
12.5
TG tok/s
5432
TTFT (ms)
31.8
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.5.1
macOS
macOS 26.5.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 188.5 | 12.5 | 31.8 GB | current |
| 4k | 154.7 | 11.8 | 33.9 GB | view |
| 8k | 130.7 | 6.6 | 38.1 GB | view |
| 16k | 86.9 | 6.5 | 45.9 GB | view |
| 32k | 99.4 | 5.6 | 57.9 GB | view |
| 64k | 88.1 | 5.8 | 85.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.5 | 1.00× |
| 2× | 21.6 | 1.73× |
| 4× | 29.1 | 2.33× |
| 8× | 27.0 | 2.16× |