Command Palette
Search for a command to run...

Tether AI Releases 460 Million Parameter Vision Language Model Optimized for Edge Devices

aiai-modelingai-model-releasesai-open-models 7 posts · 2 accounts

Tether AI released VisionPsy-Nano, an open-source vision-language artificial intelligence model with 460M parameters. The model is built to process images and text directly on phones and laptops without requiring cloud connectivity.

The company states the architecture achieved the highest overall benchmark scores in its weight class. The launch follows a separate software development kit update and a preview of the edge-optimized model posted earlier this week.

From the sources (7 posts)

@paoloardoino

Tether AI gonna release a SOTA AI model optimized for edge devices in the next few days.

@paoloardoino

Tether AI new SDK release

@tether

RT @paoloardoino: Tether AI gonna release a SOTA AI model optimized for edge devices in the next few days.

@tether

Tether Data Open-Sources VisionPsy-Nano: Best-in-Class ~460M On-Device Vision-Language Model Leading Industry Benchmarks Learn more:

@paoloardoino

Tether AI just released VisionPsy, a best-in-class state-of-the-art vision language model optimized for edge devices (phones, laptops, ...) with the highest overall score in its class/weight 🤖

@paoloardoino

Tether AI's new SOTA vision model for edge devices, live benchmarked against other models in the same class 👑

@paoloardoino

RT @qvac: Today we release VisionPsy-Nano: state-of-the-art vision-language models at 460M parameters, built to run on-device. It leads ev…

Preview built on a synthetic news corpus (16 weeks, Apr–Jul 2026). Impact calls are model reads, not price data.

About Archive