← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
32k
tokens
3,172
PP tok/s
43.7
TG tok/s
10331
TTFT (ms)
22.9
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.3.6
macOS
macOS 26.4.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,361 | 117.8 | 20.0 GB | view |
| 4k | 1,691 | 111.8 | 20.8 GB | view |
| 8k | 1,711 | 96.0 | 21.1 GB | view |
| 16k | 3,446 | 42.8 | 22.3 GB | view |
| 32k | 3,172 | 43.7 | 22.9 GB | current |
| 64k | 2,454 | 42.4 | 24.8 GB | view |
| 128k | 1,770 | 40.1 | 27.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 117.8 | 1.00× |
| 2× | 189.6 | 1.61× |
| 4× | 265.7 | 2.26× |