← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
32k
tokens
226.1
PP tok/s
22.4
TG tok/s
144917
TTFT (ms)
35.8
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
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 | 236.3 | 26.8 | 28.8 GB | view |
| 4k | 246.6 | 26.7 | 30.3 GB | view |
| 8k | 245.2 | 26.0 | 31.3 GB | view |
| 16k | 237.9 | 24.4 | 32.8 GB | view |
| 32k | 226.1 | 22.4 | 35.8 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 26.8 | 1.00× |
| 2× | 31.2 | 1.16× |
| 4× | 45.3 | 1.69× |