← Back to community benchmarks
Qwen3.5-122B-A10B-Text-qx85
Performance
8k
tokens
1,258
PP tok/s
49.7
TG tok/s
6511
TTFT (ms)
89.2
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.19
macOS
macOS 26.3.1
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 996.0 | 56.3 | 87.1 GB | view |
| 4k | 1,399 | 53.3 | 88.6 GB | view |
| 8k | 1,258 | 49.7 | 89.2 GB | current |
| 16k | 1,104 | 43.4 | 90.4 GB | view |
| 32k | 882.0 | 25.8 | 92.8 GB | view |
| 64k | 640.9 | 18.9 | 97.6 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 56.3 | 1.00× |
| 2× | 43.4 | 0.77× |
| 4× | 40.1 | 0.71× |
| 8× | 60.3 | 1.07× |