← Back to community benchmarks
Huihui-gemma-4-26B-A4B-it-abliteratedoQ8
Performance
64k
tokens
2,230
PP tok/s
50.2
TG tok/s
29388
TTFT (ms)
30.0
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,698 | 87.3 | 25.8 GB | view |
| 4k | 3,146 | 84.7 | 26.5 GB | view |
| 8k | 3,473 | 81.6 | 26.7 GB | view |
| 16k | 3,348 | 75.7 | 27.1 GB | view |
| 32k | 2,898 | 64.9 | 28.0 GB | view |
| 64k | 2,230 | 50.2 | 30.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 87.3 | 1.00× |
| 2× | 126.8 | 1.45× |
| 4× | 174.2 | 2.00× |
| 8× | 344.2 | 3.94× |