← Back to community benchmarks

Qwen3-Coder-Next

M2 Max (38c) · 96 GB · 6bit · 2026-03-08
Performance
16k
tokens
412.9
PP tok/s
38.1
TG tok/s
39676
TTFT (ms)
62.4
Peak mem (GB)
Hardware
Chip M2 Max (38c)
Memory 96 GB
GPU Cores 38
Software
oMLX v0.2.6
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 464.3 53.4 61.6 GB view
4k 443.4 50.2 61.7 GB view
8k 427.1 46.1 62.0 GB view
16k 412.9 38.1 62.4 GB current
32k 379.4 25.5 63.3 GB view
Batching Results
Batch Size TG tok/s Speedup
53.4 1.00×
65.1 1.22×
78.2 1.46×
97.9 1.83×