← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
16k
tokens
677.1
PP tok/s
43.8
TG tok/s
24198
TTFT (ms)
67.7
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 26.3
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 712.9 | 53.7 | 66.3 GB | view |
| 4k | 737.7 | 51.1 | 66.7 GB | view |
| 8k | 721.6 | 49.2 | 67.0 GB | view |
| 16k | 677.1 | 43.8 | 67.7 GB | current |
| 32k | 597.5 | 37.5 | 69.1 GB | view |
| 64k | 473.1 | 29.1 | 71.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 53.7 | 1.00× |
| 2× | 87.4 | 1.63× |
| 4× | 127.5 | 2.37× |
| 8× | 167.6 | 3.12× |