← Back to community benchmarks
Qwen3.6-35B-A3B-UD
Performance
32k
tokens
583.6
PP tok/s
38.5
TG tok/s
56144
TTFT (ms)
23.5
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
48 GB
GPU Cores
16
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 655.9 | 65.0 | 20.4 GB | view |
| 4k | 748.7 | 62.9 | 21.2 GB | view |
| 8k | 733.5 | 58.5 | 21.5 GB | view |
| 16k | 667.0 | 49.5 | 22.2 GB | view |
| 32k | 583.6 | 38.5 | 23.5 GB | current |
| 64k | 444.3 | 21.5 | 26.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 65.0 | 1.00× |
| 2× | 96.6 | 1.49× |
| 4× | 119.4 | 1.84× |