← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
32k
tokens
599.6
PP tok/s
23.7
TG tok/s
54651
TTFT (ms)
129.0
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.2.24
macOS
macOS 26.4
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 676.4 | 36.8 | 123.3 GB | view |
| 4k | 742.3 | 35.4 | 124.8 GB | view |
| 8k | 728.0 | 33.7 | 125.4 GB | view |
| 16k | 683.8 | 29.1 | 126.5 GB | view |
| 32k | 599.6 | 23.7 | 129.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 36.8 | 1.00× |
| 2× | 57.8 | 1.57× |
| 4× | 82.5 | 2.24× |