← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
32k
tokens
1,960
PP tok/s
91.9
TG tok/s
16720
TTFT (ms)
21.9
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.5.0rc1
macOS
macOS 26.4
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 3,565 | 106.3 | 20.4 GB | view |
| 4k | 4,459 | 108.1 | 21.3 GB | view |
| 8k | 4,477 | 103.8 | 21.6 GB | view |
| 16k | 2,964 | 99.8 | 21.6 GB | view |
| 32k | 1,960 | 91.9 | 21.9 GB | current |
| 64k | 1,175 | 53.2 | 22.8 GB | view |
| 128k | 547.3 | 44.9 | 24.1 GB | view |
| 195k | 207.5 | 34.1 | 54.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 106.3 | 1.00× |
| 2× | 158.3 | 1.49× |