← Back to community benchmarks
Qwen3.5-122B-A10B-5bit
Performance
4k
tokens
708.1
PP tok/s
47.8
TG tok/s
5785
TTFT (ms)
80.9
Peak mem (GB)
Hardware
Chip
M3 Ultra (60c)
Memory
256 GB
GPU Cores
60
Software
oMLX
v0.2.7
macOS
macOS 26.3
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 678.6 | 49.7 | 80.6 GB | view |
| 4k | 708.1 | 47.8 | 80.9 GB | current |
| 8k | 695.4 | 42.8 | 81.2 GB | view |
| 16k | 654.9 | 37.2 | 81.9 GB | view |
| 32k | 581.1 | 24.8 | 83.3 GB | view |
| 64k | 463.4 | 17.7 | 86.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 49.7 | 1.00× |
| 2× | 73.2 | 1.47× |