← Back to community benchmarks
Qwen3.5-122B-A10B9bit
Performance
1k
tokens
644.8
PP tok/s
43.3
TG tok/s
1588
TTFT (ms)
129.5
Peak mem (GB)
Hardware
Chip
M3 Ultra (80c)
Memory
256 GB
GPU Cores
80
Software
oMLX
v0.2.9
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 644.8 | 43.3 | 129.5 GB | current |
| 4k | 848.7 | 42.3 | 129.8 GB | view |
| 8k | 867.0 | 40.8 | 130.1 GB | view |
| 16k | 828.0 | 32.1 | 130.8 GB | view |
| 32k | 736.0 | 23.6 | 132.2 GB | view |
| 64k | 571.5 | 15.1 | 134.9 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 43.3 | 1.00× |
| 2× | 68.2 | 1.58× |