← Back to community benchmarks
Qwen3.5-35B-A3B
Performance
8k
tokens
1,674
PP tok/s
74.7
TG tok/s
4895
TTFT (ms)
36.5
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,648 | 80.6 | 36.2 GB | view |
| 4k | 1,728 | 78.4 | 36.3 GB | view |
| 8k | 1,674 | 74.7 | 36.5 GB | current |
| 16k | 1,517 | 69.3 | 37.0 GB | view |
| 32k | 1,235 | 60.7 | 37.8 GB | view |
| 64k | 931.4 | 47.4 | 39.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 80.6 | 1.00× |
| 2× | 137.2 | 1.70× |
| 4× | 194.3 | 2.41× |