← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
64k
tokens
379.7
PP tok/s
22.3
TG tok/s
172607
TTFT (ms)
77.0
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.22
macOS
macOS 26.3
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 563.6 | 44.2 | 66.4 GB | view |
| 4k | 598.4 | 42.6 | 67.9 GB | view |
| 8k | 584.9 | 40.1 | 68.5 GB | view |
| 16k | 549.0 | 35.6 | 69.7 GB | view |
| 32k | 483.8 | 29.9 | 72.1 GB | view |
| 64k | 379.7 | 22.3 | 77.0 GB | current |
| 128k | 262.5 | 12.2 | 86.7 GB | view |
| 195k | 190.2 | 9.0 | 96.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 44.2 | 1.00× |
| 2× | 72.5 | 1.64× |
| 4× | 96.3 | 2.18× |