← Back to community benchmarks
Qwen3.6-27B
Performance
16k
tokens
723.5
PP tok/s
17.0
TG tok/s
22644
TTFT (ms)
32.3
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 724.0 | 18.4 | 28.3 GB | view |
| 4k | 863.0 | 18.1 | 29.8 GB | view |
| 8k | 825.6 | 17.8 | 30.8 GB | view |
| 16k | 723.5 | 17.0 | 32.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 18.4 | 1.00× |
| 2× | 32.9 | 1.79× |
| 4× | 45.8 | 2.49× |
| 8× | 54.4 | 2.96× |