← Back to community benchmarks
Qwen3.6-27B-oQ8-fp16-mtp
Performance
16k
tokens
156.2
PP tok/s
11.5
TG tok/s
104922
TTFT (ms)
32.3
Peak mem (GB)
Hardware
Chip
M2 Max (30c)
Memory
64 GB
GPU Cores
30
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 154.8 | 11.9 | 28.3 GB | view |
| 4k | 161.5 | 11.9 | 29.8 GB | view |
| 8k | 155.7 | 11.8 | 30.8 GB | view |
| 16k | 156.2 | 11.5 | 32.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 11.9 | 1.00× |
| 2× | 20.9 | 1.76× |
| 4× | 27.5 | 2.31× |