← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
8k
tokens
2,254
PP tok/s
91.7
TG tok/s
3634
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,511 | 102.0 | 19.3 GB | view |
| 4k | 2,213 | 97.4 | 20.0 GB | view |
| 8k | 2,254 | 91.7 | 20.4 GB | current |
| 16k | 2,133 | 82.7 | 21.0 GB | view |
| 32k | 1,808 | 62.7 | 22.4 GB | view |
| 64k | 1,325 | 44.8 | 25.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 102.0 | 1.00× |
| 2× | 154.9 | 1.52× |
| 4× | 207.1 | 2.03× |