← Back to community benchmarks
Qwen3.6-35B-A3B-UD
Performance
64k
tokens
2,065
PP tok/s
82.4
TG tok/s
31744
TTFT (ms)
26.2
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.4.4
macOS
macOS 26.5.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,218 | 110.1 | 20.4 GB | view |
| 4k | 3,722 | 108.3 | 21.2 GB | view |
| 8k | 3,780 | 118.5 | 21.5 GB | view |
| 16k | 3,545 | 112.0 | 22.2 GB | view |
| 32k | 2,899 | 93.7 | 23.5 GB | view |
| 64k | 2,065 | 82.4 | 26.2 GB | current |
| 128k | 1,196 | 60.9 | 31.6 GB | view |
| 195k | 842.4 | 49.1 | 37.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 110.1 | 1.00× |
| 2× | 173.2 | 1.57× |
| 4× | 247.8 | 2.25× |
| 8× | 474.4 | 4.31× |