← Back to community benchmarks
Qwen3.6-35B-A3B-oQ6
Performance
1k
tokens
2,157
PP tok/s
103.7
TG tok/s
475
TTFT (ms)
28.1
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.8rc1
macOS
macOS 26.5
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 2,157 | 103.7 | 28.1 GB | current |
| 4k | 3,614 | 94.6 | 28.9 GB | view |
| 8k | 3,863 | 95.6 | 29.2 GB | view |
| 16k | 3,734 | 89.6 | 29.9 GB | view |
| 32k | 3,191 | 75.8 | 31.2 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 103.7 | 1.00× |
| 2× | 163.3 | 1.57× |
| 4× | 242.2 | 2.34× |