← Back to community benchmarks
Qwen3.6-27B-MTPLX-Optimized-Speed
Performance
195k
tokens
177.5
PP tok/s
8.3
TG tok/s
1126852
TTFT (ms)
55.9
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4
Context
200,000
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 765.5 | 20.7 | 15.8 GB | view |
| 4k | 410.4 | 17.5 | 17.3 GB | view |
| 8k | 398.4 | 15.5 | 18.3 GB | view |
| 16k | 339.7 | 14.2 | 19.8 GB | view |
| 32k | 319.9 | 12.8 | 22.8 GB | view |
| 64k | 241.7 | 10.0 | 28.8 GB | view |
| 128k | 173.9 | 9.2 | 41.2 GB | view |
| 195k | 177.5 | 8.3 | 55.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.7 | 1.00× |
| 2× | 27.8 | 1.34× |
| 4× | 36.4 | 1.76× |
| 8× | 43.0 | 2.08× |