← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
1,578
PP tok/s
89.4
TG tok/s
10386
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 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,338 | 117.1 | 19.3 GB | view |
| 4k | 1,685 | 108.5 | 20.0 GB | view |
| 8k | 1,687 | 103.2 | 20.4 GB | view |
| 16k | 1,578 | 89.4 | 21.0 GB | current |
| 32k | 1,289 | 78.1 | 22.4 GB | view |
| 64k | 1,006 | 59.9 | 25.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 117.1 | 1.00× |
| 2× | 183.2 | 1.56× |
| 4× | 273.6 | 2.34× |