← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
64k
tokens
513.5
PP tok/s
22.6
TG tok/s
127624
TTFT (ms)
23.3
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.6
macOS
macOS 26.2
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 855.5 | 87.6 | 20.0 GB | view |
| 4k | 910.2 | 82.0 | 20.2 GB | view |
| 8k | 878.7 | 76.7 | 20.4 GB | view |
| 16k | 778.7 | 55.8 | 20.8 GB | view |
| 32k | 629.7 | 36.6 | 21.6 GB | view |
| 64k | 513.5 | 22.6 | 23.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 87.6 | 1.00× |
| 2× | 131.5 | 1.50× |
| 4× | 161.5 | 1.84× |
| 8× | 201.1 | 2.30× |