← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
4k
tokens
152.7
PP tok/s
16.2
TG tok/s
26825
TTFT (ms)
17.9
Peak mem (GB)
Hardware
Chip
M3 Max (30c)
Memory
36 GB
GPU Cores
30
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 157.9 | 16.5 | 16.4 GB | view |
| 4k | 152.7 | 16.2 | 17.9 GB | current |
| 8k | 143.0 | 15.9 | 18.9 GB | view |
| 16k | 136.7 | 10.1 | 20.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 16.5 | 1.00× |
| 2× | 13.9 | 0.84× |
| 4× | 14.6 | 0.88× |