AI SECURITY RESEARCH

IBM previews Granite models trained to distinguish structured instructions from data

By IDAKIM Digital · Published November 13, 2026

LLMON combines reserved structural tokens and explicit instruction bindings with experimental three- and eight-billion-parameter models.

IBM Granite introduced LLMON on November 12, releasing experimental Granite 4.1-derived 3B and 8B preview models and a supporting toolkit. The markup gives instructions, data and requirements explicit semantic roles, using reserved token delimiters and identifiers to bind an instruction to its intended input. IBM reports improved resistance to indirect prompt injection on static benchmark attacks while maintaining general capabilities relative to its comparison models. This is a model-training and serialization approach, not evidence that adding XML-like tags to any existing model creates the same protection. The previews include their own chat templates and conversion examples for applications already using JSON. Static benchmark resistance does not settle adaptive attacks or permission mistakes elsewhere in an agent system.

Why it matters

Evaluate the preview against a fixed set of malicious retrieved documents and ordinary tasks. Retain tool permissions and sandbox controls independently of model-level instruction handling.

Source: IBM Granite · November 12, 2026

Product claims are attributed to the linked publisher. IDAKIM Digital has not independently tested these claims.