← Back to community benchmarks
Qwen3.5-122B-A10B-Text-mxfp4
Performance
16k
tokens
489.8
PP tok/s
42.6
TG tok/s
33451
TTFT (ms)
63.3
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
16,384
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 504.7 | 57.1 | 62.0 GB | view |
| 4k | 554.7 | 54.9 | 62.3 GB | view |
| 8k | 536.4 | 47.9 | 62.6 GB | view |
| 16k | 489.8 | 42.6 | 63.3 GB | current |
| 32k | 442.0 | 27.9 | 64.7 GB | view |
| 64k | 359.4 | 16.9 | 67.5 GB | view |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 57.1 | 1.00× |
| 2× | 89.1 | 1.56× |
| 4× | 118.0 | 2.07× |
| 8× | 149.3 | 2.61× |