← Back to community benchmarks
Qwen3.6-27B-oQ6-fp16-mtp
Performance
8k
tokens
670.1
PP tok/s
20.8
TG tok/s
12226
TTFT (ms)
25.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.9rc1
macOS
macOS 26.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 756.2 | 24.2 | 22.6 GB | view |
| 4k | 704.7 | 22.9 | 24.0 GB | view |
| 8k | 670.1 | 20.8 | 25.1 GB | current |
| 16k | 611.3 | 20.8 | 26.6 GB | view |
| 32k | 551.2 | 15.0 | 29.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 24.2 | 1.00× |
| 2× | 26.0 | 1.07× |
| 4× | 29.7 | 1.23× |