← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
16k
tokens
2,032
PP tok/s
81.6
TG tok/s
8063
TTFT (ms)
21.2
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.3.0rc1
macOS
macOS 26.4
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,147 | 103.1 | 19.2 GB | view |
| 4k | 1,986 | 95.8 | 20.2 GB | view |
| 8k | 2,087 | 90.6 | 20.5 GB | view |
| 16k | 2,032 | 81.6 | 21.2 GB | current |
| 32k | 1,730 | 64.4 | 22.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 103.1 | 1.00× |
| 2× | 154.2 | 1.50× |
| 4× | 201.8 | 1.96× |