← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
2,144
PP tok/s
81.5
TG tok/s
7641
TTFT (ms)
21.0
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,536 | 98.4 | 19.3 GB | view |
| 4k | 2,213 | 94.2 | 20.0 GB | view |
| 8k | 2,274 | 89.7 | 20.4 GB | view |
| 16k | 2,144 | 81.5 | 21.0 GB | current |
| 32k | 1,813 | 62.5 | 22.4 GB | view |
| 64k | 1,308 | 44.5 | 25.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 98.4 | 1.00× |
| 2× | 150.4 | 1.53× |
| 4× | 191.0 | 1.94× |
| 8× | 232.5 | 2.36× |