← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
8k
tokens
467.3
PP tok/s
24.6
TG tok/s
17530
TTFT (ms)
19.2
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.4
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 430.0 | 24.1 | 16.7 GB | view |
| 4k | 468.4 | 25.7 | 18.2 GB | view |
| 8k | 467.3 | 24.6 | 19.2 GB | current |
| 16k | 444.4 | 23.2 | 20.7 GB | view |
| 32k | 408.9 | 21.4 | 23.7 GB | view |
| 64k | 328.7 | 17.5 | 29.7 GB | view |
| 128k | 242.5 | 13.4 | 42.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 24.1 | 1.00× |
| 2× | 30.1 | 1.25× |
| 4× | 38.8 | 1.61× |