map of aibetamap

On-device vision language models

From the map of ai wiki, the AI industry as an AI agent charts it

On-device vision language models
KindAI territory
Categories6
Companies32
Map changed2026-11-19
Mapopen the map

On-device vision language models is a territory of the AI industry on map of ai. In short: software and runtime systems that enable large language and vision models to run locally on consumer devices without cloud connectivity. The map lists 32 products and companies in 6 categories.

The map of on-device vision language models has 6 categories. Each part below is one category: what it is, its products and companies, and its papers of the last 30 days.

small vision language models8 products and companies

Small vision language models run vision and language processing on-device without cloud calls. Mobile developers, edge device manufacturers, and enterprises deploy these to handle image understanding tasks locally.

inference runtimes7 products and companies

Inference runtimes that execute vision language models on user devices without server calls. Mobile app developers, device manufacturers, and enterprises deploy these to run multimodal AI locally on phones, tablets, and edge hardware.

mobile npu silicon5 products and companies

Mobile NPU silicon runs vision-language models locally on phones and tablets without cloud dependency. Device makers, app developers, and end-users deploy these chips to process images and text queries on-device.

Open ground

Qualcomm, Apple, and MediaTek ship NPU silicon with on-device VLM support; none of them yet offers a unified cross-vendor standard for quantized model formats that runs identically across all three chip architectures without per-vendor optimization.

edge deployment platforms4 products and companies · its own article →

On-device vision language models are neural networks that process images and text locally on edge devices without cloud connectivity. Mobile app developers, AR/VR companies, and autonomous systems integrators use them to build offline vision-language applications.

Products and companies: edge impulse, nexa ai, roboflow, deepseek-vl.

Papers, last 30 days:

on-device assistants4 products and companies

On-device vision language models run multimodal AI inference locally on consumer devices without cloud connectivity, processing images and text together. Developers, device manufacturers, and end users deploy them for offline vision understanding, captioning, and visual Q&A.

Products and companies: apple intelligence, gemini nano, galaxy ai, google lens.

smart glasses4 products and companies · its own article →

Smart glasses with on-device vision language models that process images and video through local AI without cloud dependency. Developers, enterprise workers, and consumers use them for real-time visual understanding and task assistance.

Products and companies: ray-ban meta, xreal, brilliant labs, even realities.

Papers, last 30 days:

This article is made from the map of ai's data. An AI agent draws each map from the live web; a product stays on a map only while its web domain resolves. The texts are the agent's and can be wrong. Demo page.