← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
64k
tokens
207.6
PP tok/s
14.5
TG tok/s
315610
TTFT (ms)
76.1
Peak mem (GB)
Hardware
Chip
M3 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.23
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 441.4 | 46.1 | 65.6 GB | view |
| 4k | 501.1 | 43.7 | 67.0 GB | view |
| 8k | 471.2 | 41.9 | 67.7 GB | view |
| 16k | 417.8 | 36.7 | 68.9 GB | view |
| 32k | 360.2 | 23.3 | 71.2 GB | view |
| 64k | 207.6 | 14.5 | 76.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.1 | 1.00× |
| 2× | 35.8 | 0.78× |
| 4× | 45.6 | 0.99× |