← Back to community benchmarks
Qwen3.5-122B-A10B-oQ4
Performance
32k
tokens
3,444
PP tok/s
15.0
TG tok/s
9516
TTFT (ms)
73.5
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.22
macOS
macOS 26.4
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 812.7 | 14.9 | 69.9 GB | view |
| 4k | 1,025 | 14.6 | 72.9 GB | view |
| 8k | 1,027 | 14.3 | 74.1 GB | view |
| 16k | 3,688 | 15.2 | 72.2 GB | view |
| 32k | 3,444 | 15.0 | 73.5 GB | current |
| 64k | 3,009 | 14.4 | 75.3 GB | view |
| 128k | 2,363 | 14.3 | 78.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 14.9 | 1.00× |
| 2× | 27.6 | 1.85× |
| 4× | 45.7 | 3.07× |