← Back to community benchmarks
gemma-4-26b-a4b-it
Performance
32k
tokens
548.3
PP tok/s
37.2
TG tok/s
59763
TTFT (ms)
22.2
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
64 GB
GPU Cores
38
Software
oMLX
v0.3.12
macOS
macOS 26.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 542.4 | 51.8 | 20.0 GB | view |
| 4k | 594.7 | 49.2 | 20.7 GB | view |
| 8k | 602.2 | 47.3 | 20.8 GB | view |
| 16k | 586.8 | 43.3 | 21.3 GB | view |
| 32k | 548.3 | 37.2 | 22.2 GB | current |
| 64k | 484.3 | 29.4 | 24.3 GB | view |
| 128k | 382.7 | 20.6 | 28.4 GB | view |
| 195k | 303.3 | 15.5 | 33.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 51.8 | 1.00× |
| 2× | 72.5 | 1.40× |
| 4× | 88.0 | 1.70× |