← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
32k
tokens
480.4
PP tok/s
13.2
TG tok/s
68207
TTFT (ms)
36.0
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.4.0
macOS
macOS 26.5.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 664.5 | 21.4 | 28.8 GB | view |
| 4k | 709.7 | 12.6 | 30.5 GB | view |
| 8k | 662.4 | 13.7 | 31.6 GB | view |
| 16k | 541.2 | 13.8 | 33.0 GB | view |
| 32k | 480.4 | 13.2 | 36.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 21.4 | 1.00× |
| 2× | 24.8 | 1.16× |
| 4× | 35.4 | 1.65× |