Skip to content
View original post on X: OpenBMBOfficial· 44/100AI score44/100

OpenBMB says MiniCPM-V 4.7 runs on Apple Silicon via Nativ

AISummary

OpenBMB says MiniCPM-V 4.7 in its 1B and 35B-A3B sizes now supports fully on-device inference on Mac through Nativ, with Day 0 support. The company reports 24K+ prefill tokens/s and 315 tok/s decode for the 1B model, and 4K+ prefill tokens/s and 90 tok/s decode for the 35B-A3B model, tested on Nativ v0.4.x.

AIWhy it matters

The post gives concrete Apple Silicon throughput figures for both model sizes, showing what on-device multimodal inference can deliver on a Mac.

Post on XView on X
OpenBMBVerified on X
@OpenBMB

🚀 MiniCPM-V 4.7 is now running on Apple Silicon!
Thanks to our partnership with the @Nativ_AI, MiniCPM-V 4.7 (1B & 35B-A3B) now has Day 0 support for fully on-device inference on Mac.

With Nativ, you can run:
🧠 MiniCPM-V 4.7-35B-A3B — 35B capability with efficient inference
⚡ MiniCPM-V 4.7-1B — lightweight vision model for edge devices
📷 Image + video understanding
🔒 Fully local — your data stays on your device

Tested on Apple Silicon with Nativ v0.4.x:
1B: 24K+ prefill tokens/s, 315 tok/s decode
35B-A3B: 4K+ prefill tokens/s, 90 tok/s decode
Bringing powerful multimodal models closer to everyday devices. 🍎
Huge thanks to the Nativ team for making this possible!
🔗 Nativ: https://github.com/Blaizzy/nativ

Source: OpenBMB · x.comPublished · added here

Topics

On-device AIOpen-source ecosystemProduct updates