Granite 4.2 LLMs: How They’re Built

Hugging Face overview Model Architecture Pre-Training SFT: Data Preparation & Quality Control Data Quality Control SFT Training Details Phase 2 SFT for the 30B Model Reinforcement Learning: A Multi-Stage, Multi-Environment Pipeline Training Methodology The Staged Curriculum Reward Signals Foundational RL: Build the Skills Agentic RL: Learning to Act (8B / 30B) Alignment: RLHF How the Three Sizes Differ Agentic AI Infrastructure for Scalable RL Results Quantization FP8 FP4 GGUF Infrastructure Hardware Software Stack Getting Started (Transformers) Installation Basic Inference (Thinking Mode) Non-Thinking Mode Low-Effort Thinking Tool Calling Basic Tool Calling Multi-Turn with Tool Response Multi-Turn Conversations History Thinking Truncation Parsing Thinking vs. Final Answer Using with Agentic Coding Harnesses OpenCode Pi OpenHands Quantization FP8 FP4 GGUF Infrastructure Getting Started (Transformers) Installation Basic Inference (Thinking Mode) Non-Thinking Mode Low-Effort Thinking Tool Calling Basic Tool Calling Multi-Turn with Tool Response Multi-Turn Conversations History Thinking Truncation Parsing Thinking vs.

Final Answer Using with Agentic Coding Harnesses OpenCode Pi OpenHands A technical walkthrough of how we built the Granite 4.2 reasoning model family. TL;DR: Granite 4.2 is our first family of dense, decoder-only reasoning LLMs, released in three sizes: 3B, 8B, and 30B. Each model is pre-trained from scratch on roughly 15T tokens with a five-phase strategy that extends the context window to 512K tokens, supervised fine-tuned on chain-of-thought, reasoning, and agentic-trajectory data, then post-trained with a multi-stage reinforcement learning pipeline. That pipeline includes agentic RL, where the 8B and 30B models learn to act with tools inside real sandboxed environments. Every model has a thinking / non-thinking switch, a low-effort thinking mode that spends a short reasoning budget on easy questions, and native tool calling. All Granite 4.2 models are released under the Apache 2.0 license. Granite 4.2 is the reasoning-focused release of the Granite language-model family. Earlier Granite releases were strong instruction-following assistants; Granite 4.2 adds explicit reasoning. Every model can produce a chain of thought before its answer and can run in thinking or non-thinking mode depending on how much deliberation a task needs. A low-effort mode falls between the two, spending a short reasoning budget on easy questions. The three sizes (3B, 8B, and 30B) share the same architectural design and follow the same training pipeline (pre-training from scratch, SFT, then multi-stage RL), each at its own scale. All three are strong reasoners and instruction followers. The clearest capability split shows up in post-training. The 8B and 30B models additionally go through an agentic RL block that teaches them to operate as agents: calling tools, editing and running code, driving a terminal, and searching the web inside real environments. Every model supports native tool calling.