← Back to community benchmarks
gemma-4-26b-a4b-it-oQ8
Performance
32k
tokens
1,103
PP tok/s
30.5
TG tok/s
29704
TTFT (ms)
27.5
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,132 | 77.0 | 25.7 GB | view |
| 4k | 1,307 | 73.4 | 26.2 GB | view |
| 8k | 1,386 | 66.7 | 26.3 GB | view |
| 16k | 1,306 | 44.9 | 26.7 GB | view |
| 32k | 1,103 | 30.5 | 27.5 GB | current |
| 64k | 934.9 | 13.9 | 29.2 GB | view |
| 128k | 609.1 | 7.5 | 32.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 77.0 | 1.00× |
| 2× | 115.3 | 1.50× |
| 4× | 152.3 | 1.98× |