← Back to community benchmarks
gemma-4-31b-it-mxfp8
Performance
1k
tokens
187.8
PP tok/s
11.8
TG tok/s
5451
TTFT (ms)
32.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.1
macOS
macOS 15.7.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 187.8 | 11.8 | 32.3 GB | current |
| 4k | 187.5 | 11.4 | 33.9 GB | view |
| 8k | 184.2 | 10.8 | 34.5 GB | view |
| 16k | 178.8 | 8.9 | 35.6 GB | view |
| 32k | 167.3 | 4.8 | 37.9 GB | view |
| 64k | 148.3 | 2.6 | 42.6 GB | view |
| 128k | 115.0 | 0.5 | 52.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 11.8 | 1.00× |
| 2× | 17.4 | 1.47× |