← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
16k
tokens
1,544
PP tok/s
69.4
TG tok/s
10609
TTFT (ms)
37.1
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8.dev3
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,292 | 89.2 | 35.4 GB | view |
| 4k | 1,651 | 85.2 | 36.2 GB | view |
| 8k | 1,681 | 79.9 | 36.5 GB | view |
| 16k | 1,544 | 69.4 | 37.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 89.2 | 1.00× |
| 2× | 141.3 | 1.58× |
| 4× | 210.1 | 2.36× |
| 8× | 265.3 | 2.97× |