← Back to community benchmarks
Qwen3.6-27B-mtp
Performance
8k
tokens
207.1
PP tok/s
12.3
TG tok/s
39552
TTFT (ms)
30.8
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 204.8 | 12.7 | 28.3 GB | view |
| 4k | 213.0 | 12.6 | 29.8 GB | view |
| 8k | 207.1 | 12.3 | 30.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 12.7 | 1.00× |
| 2× | 24.1 | 1.90× |
| 4× | 35.1 | 2.76× |