ZeticMLange 1.9.0 adds on-device RAG and VLM
ZeticMLange 1.9.0 introduces major capabilities for local, on-device AI operations across Android, iOS, and Flutter. The update adds support for vision-language models (LFM-VL), allowing developers to run image and text processing locally, as well as on-device Retrieval-Augmented Generation (RAG) and function calling, ensuring high privacy, reduced latency, and lower cloud computing costs by leveraging local mobile NPUs.
Running vision-language models and RAG directly on mobile devices is a game-changer for privacy and latency, but hardware fragmentation remains a major hurdle.
* Vision-language support allows local text-and-image reasoning without network dependencies.
* On-device RAG and function calling enable offline, private agentic workflows.
* Performance will vary drastically across different generations of Android and iOS NPUs.
DISCOVERED
1h ago
2026-07-06
PUBLISHED
1h ago
2026-07-06
RELEVANCE
AUTHOR
zeticai_