← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
32k
tokens
396.6
PP tok/s
23.7
TG tok/s
82618
TTFT (ms)
97.5
Peak mem (GB)
Hardware
Chip
M2 Ultra (76c)
Memory
192 GB
GPU Cores
76
Software
oMLX
v0.2.13
macOS
macOS 26.3.1
Context
32,768
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 411.8 | 40.7 | 94.8 GB | view |
| 4k | 442.0 | 39.2 | 95.1 GB | view |
| 32k | 396.6 | 23.7 | 97.5 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 40.7 | 1.00× |
| 2× | 63.6 | 1.56× |
| 4× | 81.9 | 2.01× |
| 8× | 102.0 | 2.51× |