← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
1k
tokens
239.8
PP tok/s
39.3
TG tok/s
4270
TTFT (ms)
16.7
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 239.8 | 39.3 | 16.7 GB | current |
| 4k | 236.1 | 38.3 | 18.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 39.3 | 1.00× |
| 2× | 48.1 | 1.22× |
| 4× | 52.0 | 1.32× |