← Back to community benchmarks
gemma-4-26B-A4B-it-oQ8
Performance
64k
tokens
676.9
PP tok/s
27.7
TG tok/s
96817
TTFT (ms)
29.2
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
96 GB
GPU Cores
38
Software
oMLX
v0.4.1
macOS
macOS 15.7.5
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 881.1 | 51.7 | 25.8 GB | view |
| 4k | 1,004 | 44.2 | 26.2 GB | view |
| 8k | 972.9 | 42.9 | 26.4 GB | view |
| 16k | 944.3 | 39.0 | 26.7 GB | view |
| 32k | 827.2 | 32.7 | 27.5 GB | view |
| 64k | 676.9 | 27.7 | 29.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 51.7 | 1.00× |
| 2× | 60.4 | 1.17× |
| 4× | 83.2 | 1.61× |
| 8× | 166.0 | 3.21× |