ModelRefs / Instruction-Tuned Model — AI Glossary
Instruction-Tuned Model — AI Glossary
A base model further trained to follow natural-language instructions via supervised fine-tuning and preference alignment. Also called chat model or RLHF model.
Overview
Instruction tuning (SFT on human-written input-output pairs, then RLHF or DPO) transforms a next-token predictor into a helpful assistant. GPT-4, Claude, Gemini, and Llama-Instruct are all instruction-tuned.
Reference details
| Topic | models |
|---|---|
| Also known as | chat model, RLHF model, aligned model |
| Last reviewed | 2026-06-24 |
Related terms
Commonly confused with
An instruction-tuned model follows directions; a base model of the same weights does not. It is also not the same as a reasoning model, which is trained to deliberate at length before answering. Chat model is a near-synonym in practice, usually implying the multi-turn message format on top of instruction tuning. Check which variant a model name refers to before comparing results.
Continue your research
Use these connected ModelRefs sections to compare alternatives, inspect implementation paths, and review the evidence and governance boundaries relevant to Instruction-Tuned Model — AI Glossary.
Frequently asked questions
What is Instruction-Tuned Model?
A base model further trained to follow natural-language instructions via supervised fine-tuning and preference alignment.
Is Instruction-Tuned Model the same as chat model?
Yes — chat model, RLHF model, aligned model are common aliases for Instruction-Tuned Model.
What concepts are related to Instruction-Tuned Model?
Closely related concepts include base model, sft, rlhf, dpo.