← Back to community benchmarks
Qwen3.6-27B
Performance
4k
tokens
101.5
PP tok/s
9.9
TG tok/s
40371
TTFT (ms)
34.6
Peak mem (GB)
Hardware
Chip
M2 Max (30c)
Memory
64 GB
GPU Cores
30
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 99.2 | 10.1 | 33.0 GB | view |
| 4k | 101.5 | 9.9 | 34.6 GB | current |
| 8k | 102.0 | 9.8 | 35.6 GB | view |
| 16k | 99.9 | 9.4 | 37.1 GB | view |
| 32k | 95.4 | 8.8 | 40.1 GB | view |
| 64k | 87.1 | 7.4 | 46.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 10.1 | 1.00× |
| 2× | 10.2 | 1.01× |
| 4× | 17.5 | 1.73× |