← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
32k
tokens
3,369
PP tok/s
91.8
TG tok/s
9728
TTFT (ms)
31.0
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.5.3
macOS
macOS 26.5.2
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 3,574 | 101.1 | 27.4 GB | view |
| 4k | 4,366 | 96.4 | 28.3 GB | view |
| 8k | 4,409 | 98.7 | 28.8 GB | view |
| 16k | 4,002 | 99.1 | 29.5 GB | view |
| 32k | 3,369 | 91.8 | 31.0 GB | current |
| 64k | 2,417 | 78.3 | 33.9 GB | view |
| 128k | 1,525 | 59.5 | 39.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 101.1 | 1.00× |
| 2× | 187.7 | 1.86× |
| 4× | 277.7 | 2.75× |