← Back to community benchmarks
gemma-4-26B-A4B-it-oQ8
Performance
32k
tokens
1,216
PP tok/s
60.4
TG tok/s
26955
TTFT (ms)
28.0
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 15.7.4
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,175 | 76.7 | 25.8 GB | view |
| 4k | 1,388 | 74.5 | 26.3 GB | view |
| 8k | 1,397 | 72.2 | 26.8 GB | view |
| 16k | 1,341 | 68.1 | 27.1 GB | view |
| 32k | 1,216 | 60.4 | 28.0 GB | current |
| 64k | 1,011 | 49.9 | 30.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 76.7 | 1.00× |
| 2× | 111.0 | 1.45× |
| 4× | 157.3 | 2.05× |