← Back to community benchmarks
Qwen3.6-27B-oQ6-mtp
User commentQwen3.6-27B-oQ6 model without MTP Optimizations
Performance
16k
tokens
127.1
PP tok/s
8.9
TG tok/s
128949
TTFT (ms)
26.6
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
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 233.9 | 19.7 | 22.6 GB | view |
| 4k | 191.8 | 17.4 | 24.0 GB | view |
| 8k | 180.2 | 18.3 | 25.1 GB | view |
| 16k | 127.1 | 8.9 | 26.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 19.7 | 1.00× |
| 2× | 17.8 | 0.90× |
| 4× | 23.4 | 1.19× |