This Hugging Face Space, created by Google, helps users understand and manage token usage for vision models. Users upload an image or select a sample, then the tool creates resized versions that fit preset token budgets (70, 140, 280, 560, 1120 tokens). It displays the new dimensions and preserves the original aspect ratio.
In the AI & ML space, Gemma 4 takes a focused approach. It focuses on optimizing images to fit within specific token budgets for vision-language models without losing important visual information. It is built as a consumer product for AI developers and researchers. Gemma 4 is free to use. Gemma 4 is available on the web.
Behind Gemma 4 is Google, and it first shipped in 2025. Among its 4 catalogued features are Image Resizing, Token Budget Presets, and Aspect Ratio Preservation.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do