← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
8k
tokens
2,401
PP tok/s
89.7
TG tok/s
3413
TTFT (ms)
20.5
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.5.1
macOS
macOS 26.5.2
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,079 | 96.1 | 19.3 GB | view |
| 4k | 2,485 | 93.4 | 20.1 GB | view |
| 8k | 2,401 | 89.7 | 20.5 GB | current |
| 16k | 2,188 | 83.5 | 21.3 GB | view |
| 32k | 1,831 | 73.5 | 23.3 GB | view |
| 64k | 1,336 | 59.4 | 26.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 96.1 | 1.00× |
| 2× | 155.7 | 1.62× |
| 4× | 221.5 | 2.30× |
| 8× | 442.6 | 4.61× |