---
schema: 1
type: term
title: _post-training_
description: >-
  _Post-training_ is the stage of model development that comes after a base
  model is pretrained, where the raw foundation model is shaped into a more
  capable, purpose-built system.
canonical: 'https://liquidgold.cc/channel/everyinc/explore/term/post-training-2fa987de5c4d'
markdown: >-
  https://liquidgold.cc/channel/everyinc/explore/term/post-training-2fa987de5c4d.md
llms: null
parent: 'https://liquidgold.cc/channel/everyinc/llms.txt'
channel: everyinc
updated: '2026-10-23T22:12:35.346Z'
language: null
children: []
---


# \_post-training\_

Canonical: https://liquidgold.cc/channel/everyinc/explore/term/post-training-2fa987de5c4d

Parent: [Channel index](https://liquidgold.cc/channel/everyinc/llms.txt)

## Definition

_Post-training_ is the stage of model development that comes after a base model is pretrained, where the raw foundation model is shaped into a more capable, purpose-built system.

In this channel's material, **_post-training_** is the stage of model development that comes *after* a base model is pretrained — the work of shaping a raw foundation model into something useful for a specific purpose, such as acting as a coding agent.