← Back to community benchmarks
Qwen3-Coder-Next-mxfp4
Performance
64k
tokens
298.4
PP tok/s
18.3
TG tok/s
219591
TTFT (ms)
47.0
Peak mem (GB)
Hardware
Chip
M1 Max (32c)
Memory
64 GB
GPU Cores
32
Software
oMLX
v0.2.18
macOS
macOS 26.3.1
Context
65,536
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 391.2 | 46.7 | 40.8 GB | view |
| 4k | 454.5 | 45.3 | 41.7 GB | view |
| 8k | 437.2 | 42.5 | 42.1 GB | view |
| 16k | 414.9 | 35.2 | 42.8 GB | view |
| 32k | 371.8 | 26.9 | 44.2 GB | view |
| 64k | 298.4 | 18.3 | 47.0 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 46.7 | 1.00× |
| 2× | 67.5 | 1.45× |
| 4× | 86.4 | 1.85× |
| 8× | 104.8 | 2.24× |