← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
8k
tokens
562.6
PP tok/s
41.9
TG tok/s
14560
TTFT (ms)
96.1
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.0
macOS
macOS 26.4
Context
8,192
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 497.6 | 46.1 | 94.0 GB | view |
| 4k | 567.6 | 44.3 | 95.5 GB | view |
| 8k | 562.6 | 41.9 | 96.1 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.1 | 1.00× |
| 2× | 64.7 | 1.40× |
| 4× | 75.6 | 1.64× |