← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
8k
tokens
824.8
PP tok/s
15.0
TG tok/s
9932
TTFT (ms)
30.8
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 723.5 | 18.2 | 28.3 GB | view |
| 4k | 865.8 | 17.9 | 29.8 GB | view |
| 8k | 824.8 | 15.0 | 30.8 GB | current |
| 16k | 735.5 | 16.9 | 32.3 GB | view |
| 128k | 435.2 | 8.1 | 49.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 18.2 | 1.00× |
| 2× | 32.7 | 1.80× |
| 4× | 46.0 | 2.53× |