← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
16k
tokens
1,276
PP tok/s
63.6
TG tok/s
12836
TTFT (ms)
37.0
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.8
macOS
macOS 26.2
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,635 | 83.3 | 36.2 GB | view |
| 4k | 1,722 | 76.5 | 36.3 GB | view |
| 8k | 1,494 | 76.0 | 36.5 GB | view |
| 16k | 1,276 | 63.6 | 37.0 GB | current |
| 32k | 1,053 | 44.5 | 37.8 GB | view |
| 64k | 806.8 | 27.1 | 39.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 83.3 | 1.00× |
| 2× | 126.6 | 1.52× |
| 4× | 129.0 | 1.55× |
| 8× | 185.9 | 2.23× |