← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
4k
tokens
2,101
PP tok/s
64.6
TG tok/s
1950
TTFT (ms)
36.2
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,376 | 66.7 | 35.4 GB | view |
| 4k | 2,101 | 64.6 | 36.2 GB | current |
| 8k | 2,145 | 61.8 | 36.5 GB | view |
| 16k | 1,988 | 56.4 | 37.1 GB | view |
| 32k | 1,738 | 46.9 | 38.5 GB | view |
| 64k | 1,265 | 34.6 | 41.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 66.7 | 1.00× |
| 2× | 101.1 | 1.52× |
| 4× | 139.6 | 2.09× |