← Back to community benchmarks
Qwen3.6-35B-A3B-NSC-ACE-SABER-4bit-MTPLX-Optimized-Speed
Performance
4k
tokens
2,250
PP tok/s
98.0
TG tok/s
1821
TTFT (ms)
20.0
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.8
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,545 | 101.7 | 19.3 GB | view |
| 4k | 2,250 | 98.0 | 20.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 101.7 | 1.00× |
| 2× | 136.7 | 1.34× |
| 4× | 191.9 | 1.89× |