← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
1k
tokens
716.7
PP tok/s
27.2
TG tok/s
1429
TTFT (ms)
28.8
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 716.7 | 27.2 | 28.8 GB | current |
| 4k | 777.2 | 28.3 | 30.3 GB | view |
| 8k | 750.3 | 28.3 | 31.3 GB | view |
| 16k | 722.0 | 26.5 | 32.8 GB | view |
| 32k | 670.9 | 25.4 | 34.6 GB | view |
| 64k | 576.6 | 22.5 | 41.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 27.2 | 1.00× |
| 2× | 35.4 | 1.30× |
| 4× | 50.8 | 1.87× |