← Back to community benchmarks
gemma-4-26B-A4B-it-oQ6
Performance
8k
tokens
583.6
PP tok/s
33.8
TG tok/s
14038
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 638.6 | 42.1 | 20.1 GB | view |
| 4k | 630.5 | 38.7 | 20.6 GB | view |
| 8k | 583.6 | 33.8 | 20.7 GB | current |
| 16k | 536.9 | 23.9 | 21.1 GB | view |
| 32k | 479.9 | 14.5 | 21.9 GB | view |
| 64k | 397.5 | 8.2 | 23.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 42.1 | 1.00× |
| 2× | 59.7 | 1.42× |
| 4× | 62.2 | 1.48× |
| 8× | 75.0 | 1.78× |