← Back to community benchmarks
gemma-4-31b-it-oQ8
Performance
64k
tokens
194.8
PP tok/s
12.3
TG tok/s
336499
TTFT (ms)
45.1
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 248.8 | 18.6 | 31.8 GB | view |
| 4k | 253.1 | 18.3 | 34.5 GB | view |
| 8k | 251.3 | 17.9 | 35.0 GB | view |
| 16k | 241.7 | 16.9 | 36.1 GB | view |
| 32k | 224.1 | 15.2 | 39.3 GB | view |
| 64k | 194.8 | 12.3 | 45.1 GB | current |
| 128k | 151.9 | 7.3 | 57.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 18.6 | 1.00× |
| 2× | 31.5 | 1.69× |
| 4× | 39.2 | 2.11× |