← Back to community benchmarks

Qwen3-Coder-Next

M1 Max (24c) · 64 GB · 3bit · 2026-03-12
Performance
4k
tokens
318.2
PP tok/s
38.5
TG tok/s
12871
TTFT (ms)
33.9
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 64 GB
GPU Cores 24
Software
oMLX v0.2.9
macOS macOS 26.3.1
Context 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 292.6 41.1 33.8 GB view
4k 318.2 38.5 33.9 GB current
8k 293.8 35.6 34.2 GB view
Batching Results
Batch Size TG tok/s Speedup
41.1 1.00×
54.8 1.33×
60.0 1.46×
61.6 1.50×