← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
32k
tokens
481.9
PP tok/s
30.6
TG tok/s
67996
TTFT (ms)
71.2
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.23
macOS
macOS 26.4
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 523.3 | 58.2 | 65.6 GB | view |
| 4k | 584.8 | 54.1 | 67.0 GB | view |
| 8k | 579.5 | 52.2 | 67.7 GB | view |
| 16k | 546.6 | 45.8 | 68.9 GB | view |
| 32k | 481.9 | 30.6 | 71.2 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 58.2 | 1.00× |
| 2× | 87.0 | 1.49× |
| 4× | 109.8 | 1.89× |