← Back to community benchmarks
gemma-4-26B-A4B-it-uncensored-oQ5
Performance
64k
tokens
451.1
PP tok/s
30.6
TG tok/s
145280
TTFT (ms)
20.8
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 675.8 | 52.8 | 17.4 GB | view |
| 4k | 715.8 | 51.7 | 17.8 GB | view |
| 8k | 652.3 | 49.4 | 18.2 GB | view |
| 16k | 589.1 | 45.8 | 18.3 GB | view |
| 32k | 537.7 | 37.4 | 19.6 GB | view |
| 64k | 451.1 | 30.6 | 20.8 GB | current |
| 128k | 332.7 | 19.2 | 24.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 52.8 | 1.00× |
| 2× | 77.1 | 1.46× |
| 4× | 154.2 | 2.92× |