← Back to community benchmarks
Qwen3.6-35B-A3B
Performance
8k
tokens
3,965
PP tok/s
96.0
TG tok/s
2066
TTFT (ms)
36.5
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8
macOS
macOS 26.5
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,251 | 100.3 | 35.4 GB | view |
| 4k | 3,731 | 99.2 | 36.2 GB | view |
| 8k | 3,965 | 96.0 | 36.5 GB | current |
| 16k | 3,840 | 91.6 | 37.4 GB | view |
| 32k | 3,280 | 85.0 | 38.9 GB | view |
| 64k | 2,382 | 71.8 | 41.2 GB | view |
| 128k | 1,538 | 56.3 | 46.5 GB | view |
| 195k | 1,073 | 43.5 | 54.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 100.3 | 1.00× |
| 4× | 244.6 | 2.44× |
| 8× | 323.2 | 3.22× |