← Back to community benchmarks
Qwen3.5-122B-A10B-oQ4
Performance
8k
tokens
725.6
PP tok/s
44.3
TG tok/s
11289
TTFT (ms)
69.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.3.1
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 614.4 | 46.8 | 67.4 GB | view |
| 4k | 725.3 | 46.3 | 68.9 GB | view |
| 8k | 725.6 | 44.3 | 69.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.8 | 1.00× |
| 2× | 73.6 | 1.57× |
| 4× | 98.9 | 2.11× |