← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
16k
tokens
198.4
PP tok/s
22.5
TG tok/s
82596
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 241.7 | 27.6 | 16.4 GB | view |
| 4k | 216.1 | 22.5 | 17.9 GB | view |
| 8k | 209.7 | 25.3 | 18.9 GB | view |
| 16k | 198.4 | 22.5 | 20.4 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 27.6 | 1.00× |
| 2× | 38.9 | 1.41× |
| 4× | 46.9 | 1.70× |