← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
1,619
PP tok/s
93.9
TG tok/s
10118
TTFT (ms)
21.0
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.3
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,375 | 118.2 | 19.3 GB | view |
| 4k | 1,706 | 113.6 | 20.0 GB | view |
| 8k | 1,671 | 106.9 | 20.4 GB | view |
| 16k | 1,619 | 93.9 | 21.0 GB | current |
| 32k | 1,400 | 76.8 | 22.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 118.2 | 1.00× |
| 2× | 193.4 | 1.64× |
| 4× | 263.3 | 2.23× |