← Back to community benchmarks
Qwopus3.6-27B-v2-oQ8-fp16-mtp
Performance
4k
tokens
870.9
PP tok/s
18.1
TG tok/s
4703
TTFT (ms)
29.9
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.4.4
macOS
macOS 26.5.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 717.0 | 18.6 | 28.5 GB | view |
| 4k | 870.9 | 18.1 | 29.9 GB | current |
| 8k | 802.8 | 18.2 | 30.5 GB | view |
| 16k | 700.5 | 17.6 | 31.8 GB | view |
| 32k | 624.0 | 17.1 | 34.3 GB | view |
| 64k | 542.6 | 16.1 | 39.3 GB | view |
| 128k | 405.2 | 14.0 | 49.7 GB | view |
| 195k | 313.5 | 12.6 | 60.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 18.6 | 1.00× |
| 2× | 33.4 | 1.80× |
| 4× | 48.5 | 2.61× |