← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
8k
tokens
318.7
PP tok/s
21.3
TG tok/s
25706
TTFT (ms)
30.8
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 300.7 | 22.7 | 28.3 GB | view |
| 4k | 320.1 | 21.2 | 29.8 GB | view |
| 8k | 318.7 | 21.3 | 30.8 GB | current |
| 16k | 310.4 | 21.1 | 32.3 GB | view |
| 32k | 292.2 | 14.4 | 35.3 GB | view |
| 64k | 257.2 | 16.7 | 41.3 GB | view |
| 128k | 203.7 | 4.9 | 53.7 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 22.7 | 1.00× |
| 2× | 27.0 | 1.19× |
| 4× | 32.8 | 1.44× |