← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
16k
tokens
1,030
PP tok/s
71.5
TG tok/s
15913
TTFT (ms)
21.2
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.23
macOS
macOS 26.3.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 831.3 | 80.3 | 19.3 GB | view |
| 4k | 996.0 | 82.3 | 20.2 GB | view |
| 8k | 1,061 | 78.4 | 20.5 GB | view |
| 16k | 1,030 | 71.5 | 21.2 GB | current |
| 32k | 909.6 | 58.1 | 22.5 GB | view |
| 64k | 645.0 | 33.4 | 25.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 80.3 | 1.00× |
| 2× | 102.3 | 1.27× |
| 4× | 119.1 | 1.48× |