← Back to community benchmarks
Qwen3.6-27B-oQ4-fp16-mtp
Performance
32k
tokens
284.5
PP tok/s
27.7
TG tok/s
115172
TTFT (ms)
22.3
Peak mem (GB)
Hardware
Chip
M2 Ultra (60c)
Memory
64 GB
GPU Cores
60
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 284.0 | 33.6 | 16.4 GB | view |
| 4k | 307.8 | 32.5 | 17.9 GB | view |
| 32k | 284.5 | 27.7 | 22.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 33.6 | 1.00× |
| 2× | 49.4 | 1.47× |
| 4× | 56.6 | 1.68× |