← Back to community benchmarks
Qwen3.6-27B
Performance
4k
tokens
125.5
PP tok/s
10.1
TG tok/s
32647
TTFT (ms)
34.6
Peak mem (GB)
Hardware
Chip
M2 Max (38c)
Memory
64 GB
GPU Cores
38
Software
oMLX
v0.4.4
macOS
macOS 26.5
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 120.5 | 10.0 | 33.0 GB | view |
| 4k | 125.5 | 10.1 | 34.6 GB | current |
| 8k | 124.2 | 10.0 | 35.6 GB | view |
| 16k | 122.8 | 9.7 | 37.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 10.0 | 1.00× |
| 2× | 12.7 | 1.27× |
| 4× | 21.0 | 2.10× |
| 8× | 22.3 | 2.23× |