Explore LiquidFormer: Revolutionizing Language Model Efficiency

Explore LiquidFormer: Revolutionizing Language Model EfficiencyExplore LiquidFormer: Revolutionizing Language Model Efficiency

The network for creativity

Join 1.25M professional creatives like you

Connect with clients, get discovered, and run your business 100% commission-free

Creatives on Contra have earned over $150M and we are just getting started

Back to feedPost

Sai Sudan

• May 30

LiquidFormer is a next-generation language model architecture that replaces standard Feed-Forward Networks (FFN) with Liquid Time-Constant (LTC) cells. By combining the adaptive temporal dynamics of Liquid Neural Networks with the global context of Transformers, LiquidFormer achieves state-of-the-art efficiency and sequence adaptation.

Completed work

Data Analysis Python PyTorch Visual Studio Code

Sai Sudan

• May 30

An experimental, endlessly evolving Game AI. Built using Liquid Neural Networks (LNN) and Reinforcement Learning (PPO), Rathabij learns from every defeat, adapting dynamically to the player's strategy. It also features an integrated Language Foundation Model (LFM) for real-time, contextual trashtalking.

Completed work

Data Analysis Python PyTorch Visual Studio Code

Muhammad Tariq

• May 30

This is AI editing

capcutdesignstudio

Nupur Shirwadkar

• May 30

Work Title: CAPCUT / Gourmet Typography Experience

Direction: Creative Typography / Brand Storytelling / AI Art Direction

What if CapCut was transformed into a luxury culinary world?

This concept reimagines the CAPCUT brand through cinematic typography crafted from sushi, ramen, mochi, seaweed textures, and editing-inspired visual elements. A modern Japanese model interacts naturally with the typography, blending food artistry, fashion, and creator culture into a single visual story.

The project explores how AI can turn a digital product into an immersive brand experience through bold visuals, premium aesthetics, and creative storytelling.

Created by: Nupur

Completed work