← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
8k
tokens
863.6
PP tok/s
75.5
TG tok/s
9486
TTFT (ms)
20.4
Peak mem (GB)
Hardware
Chip
M4 Pro (20c)
Memory
48 GB
GPU Cores
20
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 884.3 | 86.4 | 20.1 GB | view |
| 4k | 904.2 | 81.0 | 20.2 GB | view |
| 8k | 863.6 | 75.5 | 20.4 GB | current |
| 16k | 799.7 | 63.9 | 20.8 GB | view |
| 32k | 685.1 | 41.3 | 21.6 GB | view |
| 64k | 525.7 | 23.3 | 23.3 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 86.4 | 1.00× |
| 2× | 130.7 | 1.51× |
| 4× | 165.5 | 1.92× |
| 8× | 201.5 | 2.33× |