← Back to community benchmarks
Qwen3.6-27B-oQ8-mtp
Performance
16k
tokens
240.7
PP tok/s
24.0
TG tok/s
68079
TTFT (ms)
32.1
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.9
macOS
macOS 26.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 238.0 | 24.0 | 28.8 GB | view |
| 4k | 248.6 | 26.8 | 30.3 GB | view |
| 8k | 247.0 | 26.1 | 30.9 GB | view |
| 16k | 240.7 | 24.0 | 32.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 24.0 | 1.00× |
| 2× | 32.0 | 1.33× |
| 4× | 45.4 | 1.89× |
| 8× | 90.6 | 3.77× |