← Back to community benchmarks
Qwen3.6-27B-oQ6-mtp
Performance
8k
tokens
86.8
PP tok/s
13.8
TG tok/s
94371
TTFT (ms)
25.4
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.3.9
macOS
macOS 15.7.4
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 85.5 | 12.9 | 23.0 GB | view |
| 4k | 87.4 | 14.0 | 24.4 GB | view |
| 8k | 86.8 | 13.8 | 25.4 GB | current |
| 16k | 85.6 | 12.6 | 26.9 GB | view |
| 32k | 82.4 | 11.8 | 29.9 GB | view |
| 64k | 74.7 | 9.8 | 36.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.9 | 1.00× |
| 2× | 28.9 | 2.24× |
| 4× | 52.3 | 4.05× |