← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
1k
tokens
238.6
PP tok/s
16.6
TG tok/s
4292
TTFT (ms)
28.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 238.6 | 16.6 | 28.3 GB | current |
| 4k | 248.7 | 16.4 | 29.8 GB | view |
| 8k | 247.1 | 16.1 | 30.4 GB | view |
| 16k | 240.5 | 15.6 | 31.7 GB | view |
| 32k | 227.0 | 14.9 | 34.2 GB | view |
| 64k | 201.7 | 13.5 | 39.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 16.6 | 1.00× |
| 2× | 31.9 | 1.92× |
| 4× | 45.4 | 2.73× |
| 8× | 91.0 | 5.48× |