← Back to community benchmarks
Qwen3.6-27B-oQ5-mtp
Performance
1k
tokens
749.0
PP tok/s
39.8
TG tok/s
1367
TTFT (ms)
20.2
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 749.0 | 39.8 | 20.2 GB | current |
| 4k | 843.7 | 38.1 | 21.7 GB | view |
| 8k | 789.2 | 36.0 | 22.7 GB | view |
| 16k | 763.4 | 33.9 | 24.2 GB | view |
| 32k | 704.2 | 31.4 | 27.2 GB | view |
| 64k | 591.2 | 26.4 | 33.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 39.8 | 1.00× |
| 2× | 42.9 | 1.08× |
| 4× | 51.4 | 1.29× |