← Back to community benchmarks
Qwen3.6-27B
Performance
16k
tokens
683.7
PP tok/s
21.0
TG tok/s
23962
TTFT (ms)
25.4
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 730.5 | 24.5 | 22.1 GB | view |
| 4k | 831.3 | 24.2 | 23.5 GB | view |
| 8k | 783.9 | 22.9 | 24.2 GB | view |
| 16k | 683.7 | 21.0 | 25.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 24.5 | 1.00× |
| 2× | 37.3 | 1.52× |
| 4× | 46.9 | 1.91× |
| 8× | 43.1 | 1.76× |