← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
1k
tokens
650.3
PP tok/s
28.1
TG tok/s
1575
TTFT (ms)
229.8
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
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 650.3 | 28.1 | 229.8 GB | current |
| 4k | 809.3 | 27.3 | 230.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 28.1 | 1.00× |
| 2× | 32.1 | 1.14× |
| 4× | 50.8 | 1.81× |
| 8× | 39.5 | 1.41× |