← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
1k
tokens
432.7
PP tok/s
36.9
TG tok/s
2367
TTFT (ms)
66.3
Peak mem (GB)
Hardware
Chip
M2 Ultra (76c)
Memory
192 GB
GPU Cores
76
Software
oMLX
v0.2.6
macOS
macOS 15.0
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 432.7 | 36.9 | 66.3 GB | current |
| 4k | 456.2 | 39.2 | 66.7 GB | view |
| 8k | 453.5 | 35.6 | 67.0 GB | view |
| 16k | 438.6 | 29.7 | 67.7 GB | view |
| 32k | 403.4 | 21.6 | 69.1 GB | view |
| 64k | 323.9 | 14.8 | 71.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 36.9 | 1.00× |
| 2× | 67.5 | 1.83× |
| 4× | 95.4 | 2.59× |
| 8× | 127.7 | 3.46× |