Shakti-VLMs: Scalable Vision-Language Models for Enterprise AI

TL;DR

Shakti-VLMs are scalable vision-language models that achieve high performance on multimodal tasks with fewer data, using architectural innovations and efficient training strategies for enterprise AI applications.

Contribution

Introduction of Shakti-VLM, a family of efficient, scalable vision-language models with novel normalization and encoding techniques, optimized for data efficiency in enterprise settings.

Findings

Shakti-VLM models outperform existing models in document understanding and visual reasoning.

They achieve competitive results with significantly less training data.

Models demonstrate strong performance in OCR extraction and multimodal reasoning tasks.

Abstract

We introduce Shakti VLM, a family of vision-language models in the capacity of 1B and 4B parameters designed to address data efficiency challenges in multimodal learning. While recent VLMs achieve strong performance through extensive training data, Shakti models leverage architectural innovations to attain competitive results with fewer tokens. Key advancements include QK-Normalization for attention stability, hybrid normalization techniques, and enhanced positional encoding. A three-stage training strategy further optimizes learning efficiency. Evaluations show that Shakti-Shakti-VLM-1B and Shakti-VLM-4B excel in document understanding, Visual Reasoning, OCR extraction, and general multimodal reasoning. Our results highlight that high performance can be achieved through model design and training strategy rather than sheer data volume, making Shakti an efficient solution for enterprise-scale multimodal tasks.