← Back to community benchmarks
gemma-4-26B-A4B-it-oQ8
Performance
128k
tokens
1,702
PP tok/s
7.3
TG tok/s
76996
TTFT (ms)
32.7
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.5
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,047 | 85.8 | 25.7 GB | view |
| 4k | 3,058 | 81.5 | 26.2 GB | view |
| 8k | 3,345 | 73.9 | 26.3 GB | view |
| 16k | 3,339 | 63.1 | 26.7 GB | view |
| 32k | 3,031 | 49.0 | 27.5 GB | view |
| 64k | 2,433 | 20.8 | 29.2 GB | view |
| 128k | 1,702 | 7.3 | 32.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 85.8 | 1.00× |
| 2× | 123.7 | 1.44× |
| 4× | 175.9 | 2.05× |
| 8× | 351.4 | 4.10× |