Skip to content

EmoLLM

A mental health LLM built from instruction tuning

Visit official site

What EmoLLM is

The model is described as a mental health large language model built from LLM instruction tuning, designed to understand and support mental health at individual, group and social levels. It covers cognitive, emotional and behavioural factors, social environment, physical health, resilience, prevention, intervention and assessment tools.

What you can do with it

  • Follow an understand-support-help counselling chain
  • Use instruction tuning for psychological counselling tasks
  • Work with cognitive, emotional and behavioural health factors
  • Use the multi-turn dialogue dataset CPsyCounD and its evaluation methods

Application scenarios

  • Mental health professionals offer online counselling services
  • Researchers assess and study mental health states
  • General users learn mental health knowledge

Getting started

  • Visit the GitHub page for project background and details
  • Choose a fine-tuning configuration such as InternLM2_5_7B_chat
  • Read the quick start guide to begin using it
  • Prepare mental health datasets following the data construction guide

Pros & cons

✓ What we like

  • Covers cognitive, emotional and behavioural factors
  • Published on ModelScope and OpenXLab for access

! What to watch out for

  • Use follows a chosen fine-tuning configuration
  • Data preparation follows the data construction guide

FAQ

What is included in the model?

Cognitive, emotional and behavioural factors, social environment, physical health, resilience, prevention, intervention and assessment tools.

Where is it published?

The material names ModelScope and OpenXLab among the platforms.

Last reviewed: 2026-11-15

More AI chatbot tools

View all →

How we review