← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
64k
tokens
475.5
PP tok/s
25.8
TG tok/s
137811
TTFT (ms)
71.9
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.2.9
macOS
macOS 26.3
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 654.4 | 54.4 | 66.3 GB | view |
| 4k | 725.9 | 52.3 | 66.7 GB | view |
| 8k | 722.7 | 46.8 | 67.0 GB | view |
| 16k | 678.1 | 41.5 | 67.7 GB | view |
| 32k | 599.4 | 34.8 | 69.1 GB | view |
| 64k | 475.5 | 25.8 | 71.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 54.4 | 1.00× |
| 2× | 87.7 | 1.61× |
| 4× | 125.7 | 2.31× |
| 8× | 165.5 | 3.04× |