← Back to community benchmarks
gemma-4-26B-A4B-it-oQ5
Performance
16k
tokens
1,291
PP tok/s
78.4
TG tok/s
12695
TTFT (ms)
18.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,161 | 91.6 | 17.4 GB | view |
| 4k | 1,379 | 89.8 | 17.7 GB | view |
| 8k | 1,356 | 85.3 | 17.9 GB | view |
| 16k | 1,291 | 78.4 | 18.3 GB | current |
| 32k | 1,113 | 68.7 | 19.1 GB | view |
| 64k | 945.4 | 53.9 | 20.8 GB | view |
| 128k | 695.9 | 39.1 | 24.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 91.6 | 1.00× |
| 2× | 135.0 | 1.47× |
| 4× | 269.4 | 2.94× |