← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
16k
tokens
1,818
PP tok/s
53.2
TG tok/s
9013
TTFT (ms)
37.0
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.2.9
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,768 | 64.9 | 36.2 GB | view |
| 4k | 1,964 | 63.1 | 36.3 GB | view |
| 8k | 1,948 | 60.6 | 36.5 GB | view |
| 16k | 1,818 | 53.2 | 37.0 GB | current |
| 32k | 1,568 | 41.2 | 37.8 GB | view |
| 64k | 1,195 | 29.7 | 39.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 64.9 | 1.00× |
| 2× | 104.0 | 1.60× |
| 4× | 143.7 | 2.21× |
| 8× | 205.4 | 3.16× |