← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
32k
tokens
377.9
PP tok/s
38.6
TG tok/s
86719
TTFT (ms)
23.7
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.3.9.dev1
macOS
macOS 26.2
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 384.9 | 46.4 | 16.7 GB | view |
| 4k | 417.3 | 45.3 | 18.2 GB | view |
| 8k | 416.1 | 42.0 | 19.2 GB | view |
| 16k | 405.8 | 41.1 | 20.7 GB | view |
| 32k | 377.9 | 38.6 | 23.7 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.4 | 1.00× |
| 2× | 60.1 | 1.30× |
| 4× | 80.7 | 1.74× |
| 8× | 92.7 | 2.00× |