← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
32k
tokens
571.2
PP tok/s
16.9
TG tok/s
57372
TTFT (ms)
35.8
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 701.9 | 29.5 | 28.8 GB | view |
| 4k | 825.5 | 29.2 | 30.3 GB | view |
| 8k | 779.5 | 27.8 | 31.3 GB | view |
| 16k | 671.4 | 26.0 | 32.8 GB | view |
| 32k | 571.2 | 16.9 | 35.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 29.5 | 1.00× |
| 2× | 18.0 | 0.61× |
| 4× | 26.4 | 0.89× |