← Back to community benchmarks
nightmedia--gemma-4-26B-A4B-it-mxfp4
Performance
1k
tokens
386.6
PP tok/s
32.1
TG tok/s
2649
TTFT (ms)
14.3
Peak mem (GB)
Hardware
Chip
M4 (10c)
Memory
32 GB
GPU Cores
10
Software
oMLX
v0.3.1
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 386.6 | 32.1 | 14.3 GB | current |
| 4k | 373.7 | 31.0 | 14.7 GB | view |
| 8k | 361.8 | 3.5 | 14.8 GB | view |
| 16k | 338.1 | 21.9 | 15.2 GB | view |
| 32k | 292.6 | 10.3 | 16.0 GB | view |
| 64k | 224.6 | 3.6 | 17.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 32.1 | 1.00× |
| 4× | 47.9 | 1.49× |