← Back to community benchmarks
gemma-4-26B-A4B-it-oQ8
Performance
64k
tokens
2,289
PP tok/s
51.7
TG tok/s
28631
TTFT (ms)
32.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.12
macOS
macOS 26.6
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,906 | 78.3 | 27.9 GB | view |
| 4k | 2,988 | 75.6 | 28.4 GB | view |
| 8k | 3,224 | 73.5 | 28.8 GB | view |
| 16k | 3,172 | 72.5 | 29.2 GB | view |
| 32k | 2,884 | 62.4 | 30.1 GB | view |
| 64k | 2,289 | 51.7 | 32.1 GB | current |
| 128k | 1,628 | 37.8 | 36.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 78.3 | 1.00× |
| 2× | 118.5 | 1.51× |
| 4× | 169.1 | 2.16× |