← Back to community benchmarks
Qwen3.6-27B
Performance
16k
tokens
698.9
PP tok/s
28.5
TG tok/s
23443
TTFT (ms)
19.8
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.4.1
macOS
macOS 26.3.2
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 528.7 | 30.6 | 15.8 GB | view |
| 4k | 605.9 | 30.8 | 17.3 GB | view |
| 8k | 690.2 | 30.0 | 18.3 GB | view |
| 16k | 698.9 | 28.5 | 19.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 30.6 | 1.00× |
| 2× | 52.4 | 1.71× |
| 4× | 66.0 | 2.16× |
| 8× | 71.7 | 2.34× |