← Back to community benchmarks
Qwopus3.6-27B-v2-oQ6-fp16-mtp
Performance
1k
tokens
179.3
PP tok/s
14.6
TG tok/s
5713
TTFT (ms)
22.4
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
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 179.3 | 14.6 | 22.4 GB | current |
| 4k | 178.0 | 13.7 | 23.7 GB | view |
| 8k | 171.3 | 12.8 | 24.5 GB | view |
| 16k | 179.3 | 11.2 | 25.7 GB | view |
| 32k | 169.7 | 10.7 | 28.2 GB | view |
| 64k | 150.0 | 8.8 | 33.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 14.6 | 1.00× |
| 2× | 17.5 | 1.20× |
| 4× | 16.2 | 1.11× |
| 8× | 39.2 | 2.68× |