← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
1k
tokens
184.7
PP tok/s
29.7
TG tok/s
5543
TTFT (ms)
16.7
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
48 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 184.7 | 29.7 | 16.7 GB | current |
| 4k | 185.6 | 29.5 | 18.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 29.7 | 1.00× |
| 2× | 32.4 | 1.09× |
| 4× | 35.5 | 1.20× |