← Back to community benchmarks
Qwen3.6-27B-oQ4-fp16-mtp
Performance
8k
tokens
724.1
PP tok/s
40.1
TG tok/s
11313
TTFT (ms)
19.2
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 26.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 796.6 | 47.0 | 16.7 GB | view |
| 4k | 813.8 | 43.9 | 18.2 GB | view |
| 8k | 724.1 | 40.1 | 19.2 GB | current |
| 16k | 674.1 | 39.4 | 20.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 47.0 | 1.00× |
| 2× | 50.5 | 1.07× |
| 4× | 62.2 | 1.32× |