← Back to community benchmarks
Qwen3.5-122B-A10B-Text-mxfp4
Performance
16k
tokens
499.2
PP tok/s
40.9
TG tok/s
32820
TTFT (ms)
65.3
Peak mem (GB)
Hardware
Chip
M4 Max (40c)
Memory
128 GB
GPU Cores
40
Software
oMLX
v0.3.2
macOS
macOS 15.7.5
Context
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 471.5 | 53.2 | 62.0 GB | view |
| 4k | 531.3 | 50.2 | 63.5 GB | view |
| 8k | 521.3 | 46.9 | 64.1 GB | view |
| 16k | 499.2 | 40.9 | 65.3 GB | current |
| 32k | 442.8 | 32.3 | 67.7 GB | view |
| 64k | 350.2 | 21.1 | 72.6 GB | view |
| 128k | 241.3 | 12.4 | 82.2 GB | view |
| 195k | 173.1 | 2.4 | 92.4 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 53.2 | 1.00× |
| 2× | 83.3 | 1.57× |