← Back to community benchmarks

Qwen3-Coder-Next

M1 Ultra (48c) · 64 GB · 4bit · 2026-03-15
Performance
1k
tokens
440.0
PP tok/s
45.0
TG tok/s
2327
TTFT (ms)
43.0
Peak mem (GB)
Hardware
Chip M1 Ultra (48c)
Memory 64 GB
GPU Cores 48
Software
oMLX v0.2.11
macOS macOS 26.4
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 440.0 45.0 43.0 GB current
4k 601.7 43.4 43.2 GB view
8k 615.3 39.8 43.4 GB view
16k 599.4 36.0 43.9 GB view
32k 552.1 29.3 44.7 GB view
64k 451.8 19.9 46.5 GB view
128k 311.0 6.2 50.1 GB view
195k 223.9 2.5 54.0 GB view
Batching Results
Batch Size TG tok/s Speedup
45.0 1.00×
71.7 1.59×
103.5 2.30×
134.5 2.99×