← Back to community benchmarks
Qwen3.6-27B-mtp
Performance
8k
tokens
826.8
PP tok/s
42.1
TG tok/s
9908
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 795.1 | 46.8 | 18.3 GB | view |
| 4k | 892.8 | 46.3 | 19.7 GB | view |
| 8k | 826.8 | 42.1 | 20.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.8 | 1.00× |
| 2× | 55.2 | 1.18× |
| 4× | 54.7 | 1.17× |