← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
1k
tokens
426.8
PP tok/s
36.3
TG tok/s
2399
TTFT (ms)
94.0
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.4
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 426.8 | 36.3 | 94.0 GB | current |
| 4k | 486.3 | 35.3 | 95.3 GB | view |
| 8k | 475.9 | 33.5 | 95.9 GB | view |
| 16k | 420.5 | 30.3 | 97.1 GB | view |
| 32k | 394.8 | 24.5 | 99.5 GB | view |
| 64k | 292.8 | 18.1 | 104.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 36.3 | 1.00× |
| 2× | 55.7 | 1.53× |
| 4× | 65.2 | 1.80× |