Granite 4.0 3B Vision
granite-4.0-3b-vision
Open SourceMultimodal
About
IBM Granite 4.0 3B Vision is a vision-language model (VLM) for enterprise-grade document data extraction. Delivered as a LoRA adapter on top of Granite 4.0 Micro, allowing a single deployment to support both multimodal document understanding and text-only workloads. Integrates with Docling for document conversion pipelines. Apache 2.0.
Granite 4.0 3B Vision has a 128K-token context window.
Capabilities
VisionMultimodalReasoningFunction CallingTool UseStructured OutputsCode Execution
Specifications
FamilyGranite Vision
Released2026-04-01
Parameters3B
Context128K
ArchitectureLoRA adapter on Granite 4.0 Micro (3B dense)
Created by
Creating reliable and adaptable AI solutions
Armonk, New York, United States
Founded 1945
Website