← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
8k
tokens
2,277
PP tok/s
91.0
TG tok/s
3598
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
48 GB
GPU Cores
20
Software
oMLX
v0.3.6
macOS
macOS 26.4
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,257 | 101.7 | 19.3 GB | view |
| 4k | 2,101 | 96.9 | 20.0 GB | view |
| 8k | 2,277 | 91.0 | 20.4 GB | current |
| 16k | 2,105 | 82.4 | 21.0 GB | view |
| 32k | 1,833 | 61.8 | 22.4 GB | view |
| 64k | 1,342 | 44.7 | 25.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 101.7 | 1.00× |
| 2× | 153.3 | 1.51× |
| 4× | 201.3 | 1.98× |