← Back to community benchmarks

Nemotron-Cascade-2-30B-A3B

M5 Pro (20c) · 64 GB · 4bit · 2026-03-26
Performance
32k
tokens
1,669
PP tok/s
78.0
TG tok/s
19634
TTFT (ms)
20.0
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.2.21
macOS macOS 26.4
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,518 107.2 18.8 GB view
4k 1,868 103.6 19.9 GB view
8k 1,912 98.8 19.9 GB view
16k 1,847 91.9 19.9 GB view
32k 1,669 78.0 20.0 GB current
Batching Results
Batch Size TG tok/s Speedup
107.2 1.00×
152.8 1.43×
191.8 1.79×