← Back to community benchmarks
Qwopus3.6-27B-v2-oQ6-fp16-mtp
Performance
32k
tokens
169.7
PP tok/s
14.1
TG tok/s
193054
TTFT (ms)
28.6
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
96 GB
GPU Cores
38
Software
oMLX
v0.4.1
macOS
macOS 15.7.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 177.7 | 19.4 | 22.8 GB | view |
| 4k | 183.6 | 18.1 | 24.2 GB | view |
| 8k | 171.5 | 17.5 | 24.8 GB | view |
| 16k | 178.9 | 15.2 | 26.1 GB | view |
| 32k | 169.7 | 14.1 | 28.6 GB | current |
| 64k | 151.0 | 7.6 | 33.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.4 | 1.00× |
| 2× | 13.6 | 0.70× |
| 4× | 17.2 | 0.89× |
| 8× | 35.1 | 1.81× |