← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
1k
tokens
557.1
PP tok/s
55.9
TG tok/s
1838
TTFT (ms)
66.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.8
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 557.1 | 55.9 | 66.3 GB | current |
| 4k | 575.7 | 50.2 | 66.7 GB | view |
| 8k | 535.9 | 45.7 | 67.0 GB | view |
| 16k | 493.1 | 44.3 | 67.7 GB | view |
| 32k | 433.2 | 26.0 | 69.1 GB | view |
| 64k | 356.5 | 16.8 | 71.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 55.9 | 1.00× |
| 2× | 85.7 | 1.53× |
| 4× | 109.0 | 1.95× |
| 8× | 134.4 | 2.40× |