← Back to community benchmarks
Qwen3.5-122B-A10B
Performance
1k
tokens
244.4
PP tok/s
20.4
TG tok/s
4190
TTFT (ms)
94.8
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.2.8
macOS
macOS 26.3.1
Context
1,024
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 244.4 | 20.4 | 94.8 GB | current |
| 4k | 255.2 | 19.3 | 95.1 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 20.4 | 1.00× |
| 2× | 24.2 | 1.19× |
| 4× | 29.8 | 1.46× |