← Back to community benchmarks
Qwen3.6-27B-OptiQ
Performance
8k
tokens
247.7
PP tok/s
24.8
TG tok/s
33072
TTFT (ms)
21.7
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.4.4rc1
macOS
macOS 27.0
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 235.1 | 25.5 | 19.2 GB | view |
| 4k | 246.4 | 25.2 | 20.6 GB | view |
| 8k | 247.7 | 24.8 | 21.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 25.5 | 1.00× |
| 2× | 46.1 | 1.81× |
| 4× | 57.4 | 2.25× |
| 8× | 60.9 | 2.39× |