Tether Data Releases VisionPsy-Nano Vision Language Model
Tether Data announced in a press release the open source release of VisionPsy-Nano, a 460 million parameter vision language model designed for on device and edge deployment.
The model was developed by QVAC, Tether Data's AI research initiative. Tether Data is releasing two variants: VisionPsy-Nano-460M for quality and VisionPsy-Nano-460M-Flash for faster smartphone deployment.
Tether Data says VisionPsy-Nano-460M reached an overall normalized score of 62.3 across its benchmark tests. The company says it outperformed Liquid AI's LFM2.5-VL-450M, Hugging Face's SmolVLM2-500M, and nanoVLM-460M-8k on 16 of 17 benchmarks.
The model supports document understanding, visual perception, reasoning, and instruction following. Both variants are available under the Apache 2.0 license for research and educational use.
We hope you enjoyed this article.
Consider subscribing to one of our newsletters like Daily AI Brief.
Also, consider following us on social media:
Subscribe to Daily AI Brief
Daily report covering major AI developments and industry news, with both top stories and complete market updates
Market report
2025 State of Data Security Report: Quantifying AI’s Impact on Data Risk
The 2025 State of Data Security Report by Varonis analyzes the impact of AI on data security across 1,000 IT environments. It highlights critical vulnerabilities such as exposed sensitive cloud data, ghost users, and unsanctioned AI applications. The report emphasizes the need for robust data governance and security measures to mitigate AI-related risks.
Read more