← Back to community benchmarks
Qwen3.6-35B-A3B-oQ6
Performance
32k
tokens
1,132
PP tok/s
72.3
TG tok/s
28958
TTFT (ms)
30.8
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,342 | 92.5 | 27.7 GB | view |
| 4k | 1,653 | 87.9 | 28.5 GB | view |
| 8k | 1,559 | 85.5 | 28.9 GB | view |
| 16k | 1,339 | 79.3 | 29.7 GB | view |
| 32k | 1,132 | 72.3 | 30.8 GB | current |
| 64k | 866.7 | 59.7 | 33.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 92.5 | 1.00× |
| 2× | 137.3 | 1.48× |
| 4× | 195.8 | 2.12× |