← Back to community benchmarks
Qwen3.5-122B-A10B-oQ4
Performance
32k
tokens
3,525
PP tok/s
15.2
TG tok/s
9295
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 | 809.3 | 15.4 | 69.9 GB | view |
| 4k | 1,071 | 15.2 | 72.9 GB | view |
| 8k | 1,042 | 14.9 | 74.1 GB | view |
| 16k | 3,700 | 15.3 | 72.2 GB | view |
| 32k | 3,525 | 15.2 | 73.5 GB | current |
| 64k | 3,068 | 14.9 | 75.3 GB | view |
| 128k | 2,340 | 14.4 | 78.8 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 15.4 | 1.00× |
| 2× | 27.6 | 1.79× |
| 4× | 47.0 | 3.05× |