← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
64k
tokens
1,260
PP tok/s
45.2
TG tok/s
51997
TTFT (ms)
23.3
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.13
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,992 | 100.8 | 20.1 GB | view |
| 4k | 2,152 | 96.0 | 20.2 GB | view |
| 8k | 2,111 | 89.7 | 20.4 GB | view |
| 16k | 1,961 | 79.4 | 20.8 GB | view |
| 32k | 1,682 | 63.6 | 21.6 GB | view |
| 64k | 1,260 | 45.2 | 23.3 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 100.8 | 1.00× |
| 2× | 158.1 | 1.57× |
| 4× | 216.6 | 2.15× |