← Back to community benchmarks
gemma-4-26b-a4b-it-UD
Performance
1k
tokens
880.0
PP tok/s
77.9
TG tok/s
1164
TTFT (ms)
14.7
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 880.0 | 77.9 | 14.7 GB | current |
| 4k | 1,118 | 74.2 | 15.4 GB | view |
| 8k | 1,208 | 71.6 | 15.5 GB | view |
| 16k | 1,135 | 64.0 | 15.7 GB | view |
| 32k | 1,046 | 55.4 | 16.9 GB | view |
| 64k | 851.3 | 42.5 | 18.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 77.9 | 1.00× |
| 2× | 117.6 | 1.51× |
| 4× | 154.1 | 1.98× |
| 8× | 187.3 | 2.40× |