# Base Model

> A base model is a language model after pretraining but before instruction tuning or preference tuning, so it continues text in the style of its training data rather than following instructions.

- Identifier: PTL-0104
- Category: Foundations
- Canonical URL: https://protologue.com/t/base-model/
- Also known as: pretrained model, base LLM, completion model

## Description

Asked a question, a base model may answer it, continue with more questions, or produce something unrelated, because it only predicts plausible continuations. Prompting one effectively means framing the task as a document to complete, for example with few-shot examples or the start of the desired output. Chat assistants are derived from base models through instruction tuning and RLHF, which is why they are often called instruct or chat models.

## Example

```
A completion-style prompt for a base model: "Q: What is the capital of France?\nA: Paris\nQ: What is the capital of Japan?\nA:"
```

## Broader terms

- [Large Language Model](https://protologue.com/t/large-language-model/)

## Related terms

- [Instruction Tuning](https://protologue.com/t/instruction-tuning/)
- [Reinforcement Learning from Human Feedback](https://protologue.com/t/rlhf/)
- [Few-shot Prompting](https://protologue.com/t/few-shot-prompting/)
- [Chat Template](https://protologue.com/t/chat-template/)

## Sources

- Ouyang et al. (2022). Training language models to follow instructions with human feedback. https://arxiv.org/abs/2203.02155
- Brown et al. (2020). Language Models are Few-Shot Learners. https://arxiv.org/abs/2005.14165

## Cite this entry

Protologue. (2026). Base Model. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.4.0, PTL-0104). https://protologue.com/t/base-model/

License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/)