← Back to community benchmarks
gemma-4-26B-A4B-it-oQ4
Performance
16k
tokens
1,034
PP tok/s
59.5
TG tok/s
15842
TTFT (ms)
15.7
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
64 GB
GPU Cores
38
Software
oMLX
v0.3.9.dev2-memtest
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 936.6 | 69.8 | 14.4 GB | view |
| 4k | 1,086 | 67.2 | 15.1 GB | view |
| 8k | 1,109 | 64.6 | 15.3 GB | view |
| 16k | 1,034 | 59.5 | 15.7 GB | current |
| 32k | 880.2 | 51.8 | 16.7 GB | view |
| 64k | 665.4 | 25.7 | 18.7 GB | view |
| 128k | 419.4 | 22.1 | 22.9 GB | view |
| 195k | 289.0 | 21.9 | 27.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 69.8 | 1.00× |
| 2× | 100.8 | 1.44× |
| 4× | 125.7 | 1.80× |