← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
64k
tokens
2,265
PP tok/s
54.5
TG tok/s
28940
TTFT (ms)
73.9
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,981 | 71.9 | 67.5 GB | view |
| 4k | 3,487 | 71.1 | 68.3 GB | view |
| 8k | 3,713 | 69.5 | 68.7 GB | view |
| 16k | 3,610 | 66.7 | 69.5 GB | view |
| 32k | 3,073 | 60.9 | 70.9 GB | view |
| 64k | 2,265 | 54.5 | 73.9 GB | current |
| 128k | 1,396 | 44.7 | 80.0 GB | view |
| 195k | 959.4 | 36.5 | 86.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 71.9 | 1.00× |
| 2× | 66.2 | 0.92× |
| 4× | 112.8 | 1.57× |
| 8× | 121.8 | 1.69× |