← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
8k
tokens
929.4
PP tok/s
49.1
TG tok/s
8815
TTFT (ms)
67.0
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
512 GB
GPU Cores
80
Software
oMLX
v0.2.11
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 888.6 | 54.4 | 66.3 GB | view |
| 4k | 949.1 | 52.1 | 66.7 GB | view |
| 8k | 929.4 | 49.1 | 67.0 GB | current |
| 16k | 873.8 | 44.9 | 67.7 GB | view |
| 32k | 769.0 | 38.5 | 69.1 GB | view |
| 64k | 601.9 | 29.5 | 71.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 54.4 | 1.00× |
| 2× | 89.5 | 1.65× |
| 4× | 135.6 | 2.49× |