← Back to community benchmarks
Qwen3.6-27B
Performance
4k
tokens
312.3
PP tok/s
37.1
TG tok/s
13118
TTFT (ms)
16.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.6
macOS
macOS 15.4
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 293.8 | 39.3 | 15.1 GB | view |
| 4k | 312.3 | 37.1 | 16.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 39.3 | 1.00× |
| 2× | 66.5 | 1.69× |
| 4× | 79.4 | 2.02× |
| 8× | 92.0 | 2.34× |