Skip to content
Trending storyDeveloping

inclusionAI releases ZoomGain paper, benchmark, RL dataset and 4B/9B checkpoints

2 articles1 sourcesince Nov 10Last article

Overview

AISummary of 2 articles

inclusionAI released the ZoomGain paper, the ZoomGainBench evaluation benchmark, an RL training dataset, and ZoomGain-4B and ZoomGain-9B model checkpoints on November 11, 2026.

ZoomGain builds questions about full images from details that appear when zoomed in, using this global-to-zoom perception gap to supervise vision-language models.

The paper reports that the best of 28 models scored 61.08 on ZoomGainBench, and that ZoomGain-9B performed better than comparison models on multiple fine-grained and general perception benchmarks. Use of the Hugging Face repositories may require authentication with hf auth login, and inclusionAI directs users to each dataset and checkpoint card for its terms.

Written by AI from the articles below · updated Nov 11, 3:16 AM ET

Check the sources:

Article timeline

The articles in this story. Times are ET.

Nov 10
  1. inclusionAI (Ant Ling) · new models on Hugging FaceOfficial
    inclusionAI releases ZoomGain-9B and ZoomGain-4B vision-language models

    AIinclusionAI released the ZoomGain paper, the ZoomGainBench evaluation benchmark, an RL training dataset, and ZoomGain-4B and ZoomGain-9B model checkpoints on November 11, 2026. The paper reports that the best of 28 models scored 61.08 on ZoomGainBench, and that ZoomGain-9B reached superior performance on multiple fine-grained and general perception benchmarks.

  2. inclusionAI (Ant Ling) · new models on Hugging FaceOfficial
    inclusionAI releases ZoomGain paper, benchmark, RL data and checkpoints

    AIinclusionAI released the ZoomGain paper, the ZoomGainBench evaluation benchmark, an RL training dataset, and ZoomGain-4B and ZoomGain-9B model checkpoints on November 11, 2026. ZoomGain builds questions about full images from details that appear when zoomed in, using this global-to-zoom perception gap to supervise vision-language models. In the paper, the best of 28 models scored 61.08 on ZoomGainBench.

Heat Trend

Current comparable heat
5
Peak comparable heat
5Nov 11, 4:00 AM ET
Change in past 24 hours
–

The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.