Granite 4.0 3B Vision

Overview

IBM's 3B vision-language model purpose-built for enterprise document extraction, including OCR, table parsing, and form understanding. Apache 2.0 and laptop-deployable.

When to pick this model

• Enterprise document and form extraction pipelines
• OCR replacement for invoices, receipts, and PDFs
• Table structure understanding at scale
• Apache-licensed on-prem document AI
• Edge deployment for sensitive enterprise data

VRAM requirements by quantization

Quantization	VRAM required
Q4_K_M (recommended)	2.2 GB
Q5_K_M	2.7 GB
Q8_0	3.8 GB
FP16 (no quantization)	6.5 GB

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

Strengths

• Fast, accurate enterprise OCR
• Strong table and form-field extraction
• Apache 2.0 license
• Runs comfortably on a laptop

Limitations

• 16K context limits multi-page documents
• English-first, weak on non-Latin scripts
• Narrow scope, not a general-purpose VLM

Architecture & training

Architecture: Dense 3B VLM · specialized for enterprise documents

Training: IBM Granite 4.0 family.

Quick start

# HuggingFace : ibm-granite/granite-4.0-3b-vision

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.