← Back to community benchmarks
Qwen3.5-122B-A10B-oQ4
Performance
16k
tokens
2,100
PP tok/s
17.1
TG tok/s
7802
TTFT (ms)
86.8
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.5
macOS
macOS 26.4.1
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 1,038 | 57.2 | 83.3 GB | view |
| 4k | 1,504 | 55.3 | 84.6 GB | view |
| 8k | 1,522 | 52.6 | 85.2 GB | view |
| 16k | 2,100 | 17.1 | 86.8 GB | current |
| 32k | 1,863 | 17.0 | 88.0 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 57.2 | 1.00× |
| 2× | 82.2 | 1.44× |
| 4× | 102.5 | 1.79× |