← Back to community benchmarks
Qwen3.6-27B-oQ4-mtp
Performance
1k
tokens
194.9
PP tok/s
31.2
TG tok/s
5255
TTFT (ms)
16.7
Peak mem (GB)
Hardware
Chip
M4 Max (32c)
Memory
36 GB
GPU Cores
32
Software
oMLX
v0.3.9.dev2
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 194.9 | 31.2 | 16.7 GB | current |
| 4k | 200.2 | 31.8 | 18.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 31.2 | 1.00× |
| 2× | 39.7 | 1.27× |
| 4× | 45.1 | 1.45× |