← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
8k
tokens
718.1
PP tok/s
39.5
TG tok/s
11408
TTFT (ms)
19.2
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 767.8 | 45.2 | 16.8 GB | view |
| 4k | 767.8 | 41.9 | 18.2 GB | view |
| 8k | 718.1 | 39.5 | 19.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 45.2 | 1.00× |
| 2× | 47.5 | 1.05× |
| 4× | 60.5 | 1.34× |
| 8× | 61.9 | 1.37× |