← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
8k
tokens
3,840
PP tok/s
70.2
TG tok/s
2133
TTFT (ms)
66.7
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,020 | 73.4 | 65.6 GB | view |
| 4k | 3,581 | 71.9 | 66.4 GB | view |
| 8k | 3,840 | 70.2 | 66.7 GB | current |
| 16k | 3,735 | 67.7 | 67.6 GB | view |
| 32k | 3,226 | 64.0 | 69.1 GB | view |
| 64k | 2,310 | 54.9 | 72.0 GB | view |
| 128k | 1,469 | 46.5 | 78.1 GB | view |
| 195k | 1,076 | 38.9 | 85.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 73.4 | 1.00× |
| 2× | 66.8 | 0.91× |
| 4× | 113.4 | 1.54× |
| 8× | 127.9 | 1.74× |