← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
64k
tokens
427.6
PP tok/s
30.9
TG tok/s
153271
TTFT (ms)
23.3
Peak mem (GB)
Hardware
Chip
M4 Pro (16c)
Memory
64 GB
GPU Cores
16
Software
oMLX
v0.2.7
macOS
macOS 26.4
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 737.9 | 88.9 | 20.1 GB | view |
| 4k | 749.5 | 82.9 | 20.2 GB | view |
| 8k | 722.8 | 76.0 | 20.4 GB | view |
| 16k | 667.0 | 62.8 | 20.8 GB | view |
| 32k | 573.9 | 49.6 | 21.6 GB | view |
| 64k | 427.6 | 30.9 | 23.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 88.9 | 1.00× |
| 2× | 118.4 | 1.33× |
| 4× | 136.7 | 1.54× |
| 8× | 166.0 | 1.87× |