← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
64k
tokens
458.6
PP tok/s
15.5
TG tok/s
142920
TTFT (ms)
71.9
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
96 GB
GPU Cores
60
Software
oMLX
v0.2.6
macOS
macOS 15.6.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 638.7 | 54.4 | 66.3 GB | view |
| 4k | 714.1 | 50.9 | 66.7 GB | view |
| 8k | 712.7 | 47.3 | 67.0 GB | view |
| 16k | 670.7 | 35.9 | 67.7 GB | view |
| 32k | 587.5 | 24.4 | 69.1 GB | view |
| 64k | 458.6 | 15.5 | 71.9 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 54.4 | 1.00× |
| 2× | 87.7 | 1.61× |
| 4× | 124.9 | 2.30× |