← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
4k
tokens
4,304
PP tok/s
98.4
TG tok/s
952
TTFT (ms)
36.4
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
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 3,494 | 100.2 | 35.5 GB | view |
| 4k | 4,304 | 98.4 | 36.4 GB | current |
| 8k | 4,056 | 96.1 | 36.7 GB | view |
| 16k | 2,723 | 91.8 | 36.5 GB | view |
| 32k | 1,888 | 85.7 | 36.8 GB | view |
| 64k | 1,188 | 74.5 | 37.4 GB | view |
| 128k | 726.4 | 60.8 | 38.7 GB | view |
| 195k | 509.5 | 50.8 | 40.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 100.2 | 1.00× |
| 2× | 208.7 | 2.08× |
| 4× | 418.6 | 4.18× |
| 8× | 833.4 | 8.32× |