← Back to community benchmarks

Qwen3-Coder-Next-REAP-48B-A3B-4bit

M4 Pro (20c) · 48 GB · 4bit · 2026-04-18
Performance
1k
tokens
646.6
PP tok/s
66.5
TG tok/s
1584
TTFT (ms)
26.9
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.3.6
macOS macOS 26.4.1
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 646.6 66.5 26.9 GB current
4k 704.6 62.3 27.8 GB view
8k 688.2 55.9 28.1 GB view
16k 620.6 48.8 28.8 GB view
Batching Results
Batch Size TG tok/s Speedup
66.5 1.00×
100.4 1.51×
128.5 1.93×