← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
8k
tokens
421.2
PP tok/s
46.0
TG tok/s
19450
TTFT (ms)
21.3
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 394.6 | 50.0 | 18.9 GB | view |
| 4k | 422.3 | 47.7 | 20.3 GB | view |
| 8k | 421.2 | 46.0 | 21.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 50.0 | 1.00× |
| 2× | 62.7 | 1.25× |
| 4× | 81.1 | 1.62× |