← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
32k
tokens
3,038
PP tok/s
85.5
TG tok/s
10788
TTFT (ms)
38.5
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.4.2.dev3
macOS
macOS 26.5.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,660 | 100.1 | 35.4 GB | view |
| 4k | 3,953 | 97.7 | 36.2 GB | view |
| 8k | 3,858 | 96.8 | 36.5 GB | view |
| 16k | 3,555 | 92.9 | 37.1 GB | view |
| 32k | 3,038 | 85.5 | 38.5 GB | current |
| 64k | 2,282 | 73.7 | 41.2 GB | view |
| 128k | 1,491 | 41.1 | 46.5 GB | view |
| 195k | 1,088 | 47.8 | 52.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 100.1 | 1.00× |
| 2× | 159.8 | 1.60× |
| 4× | 321.9 | 3.22× |
| 8× | 679.5 | 6.79× |