← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
195k
tokens
898.8
PP tok/s
44.4
TG tok/s
222528
TTFT (ms)
38.6
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.4
Context
200,000
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,672 | 129.8 | 19.2 GB | view |
| 4k | 3,071 | 121.5 | 20.0 GB | view |
| 8k | 3,995 | 118.4 | 20.5 GB | view |
| 16k | 3,739 | 113.9 | 21.2 GB | view |
| 32k | 3,360 | 104.8 | 22.7 GB | view |
| 64k | 2,319 | 83.9 | 25.7 GB | view |
| 128k | 1,400 | 59.6 | 30.4 GB | view |
| 195k | 898.8 | 44.4 | 38.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 129.8 | 1.00× |
| 2× | 201.7 | 1.55× |
| 4× | 310.1 | 2.39× |
| 8× | 405.0 | 3.12× |