Ant Lingbo Open-Sources LingBot-Touch, a Tactile Foundation Model

What happened

Ant Lingbo Technology open-sourced LingBot-Touch on November 10, a tactile foundation model pretrained on nearly 5,000 hours of real tactile data spanning magnetic, vision-based, and piezoresistive sensing. In 26 real-robot tests across Sharpa Wave and XHand1 hands, four action models improved average task success rates by about 14 to 26 percentage points when using its tactile representations. The technical report, model, and code are available on Hugging Face, GitHub, and ModelScope.

Read next

• More changes in today’s briefingGoogle DeepMind Introduces Gemini 4 Hydrogen for Fast Agentic Workflows

The story so far

Ant Lingbo Technology open-sourced LingBot-Touch on November 10, a tactile foundation model pretrained on nearly 5,000 hours of real tactile data from magnetic, vision-based, and piezoresistive sensors. Its technical report, model, and code are available on Hugging Face, GitHub, and ModelScope. In 26 real-robot tests using Sharpa Wave and XHand1 hands, four action models achieved average task success rates about 14 to 26 percentage points higher when using LingBot-Touch’s tactile representations.