← Back to community benchmarks
Qwen3.6-27B-oQ4-fp16-mtp
Performance
16k
tokens
682.8
PP tok/s
38.1
TG tok/s
23997
TTFT (ms)
20.7
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 668.5 | 44.6 | 16.7 GB | view |
| 4k | 726.2 | 42.8 | 18.2 GB | view |
| 8k | 715.0 | 41.6 | 19.2 GB | view |
| 16k | 682.8 | 38.1 | 20.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 44.6 | 1.00× |
| 2× | 49.6 | 1.11× |
| 4× | 59.9 | 1.34× |