← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
4k
tokens
500.1
PP tok/s
42.5
TG tok/s
8191
TTFT (ms)
66.9
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.4
macOS
macOS 26.4
Context
4,096
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 440.1 | 44.0 | 65.6 GB | view |
| 4k | 500.1 | 42.5 | 66.9 GB | current |
| 8k | 486.1 | 40.7 | 67.5 GB | view |
| 16k | 459.3 | 35.0 | 68.7 GB | view |
| 32k | 388.6 | 29.0 | 71.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 44.0 | 1.00× |
| 2× | 68.6 | 1.56× |
| 4× | 86.8 | 1.97× |