← Back to community benchmarks
gemma-4-26B-A4B-it-oQ5
Performance
32k
tokens
3,116
PP tok/s
72.5
TG tok/s
10515
TTFT (ms)
20.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,780 | 100.1 | 17.8 GB | view |
| 4k | 3,000 | 96.3 | 18.4 GB | view |
| 8k | 3,378 | 92.7 | 18.6 GB | view |
| 16k | 3,470 | 84.2 | 19.1 GB | view |
| 32k | 3,116 | 72.5 | 20.1 GB | current |
| 64k | 2,473 | 57.5 | 22.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 100.1 | 1.00× |
| 2× | 142.7 | 1.43× |
| 4× | 184.0 | 1.84× |