ChatHaruhi_RolePlaying_qwen_7b Open-source Role-playing Dialogue Model - Supports Multi-role Scenario Conversations

Home

Chatharuhi RolePlaying Qwen 7b

Developed by silk-road

A role-playing dialogue model fine-tuned based on Qwen-7B, supporting various character settings and scenario dialogues

Large Language Model

Transformers

Open Source License:Apache-2.0 #Role-playing fine-tuning #Dialogue generation #Light novel style

Downloads 73

Release Time : 12/18/2023

Model Overview

This model is a role-playing dialogue model fine-tuned on the Qwen-7B foundation, specifically designed to generate dialogue content that aligns with specific character personalities and backgrounds. It supports loading character settings from multiple sources and can generate coherent character responses based on context.

Model Features

Multi-character support

Supports loading character settings in various ways, including preset character names, HuggingFace character libraries, or custom JSONL format character files

Context awareness

Capable of generating responses that match character personalities based on dialogue history and character background

Large-scale training data

Fine-tuned using 118K role-playing dialogue data entries to enhance the accuracy and diversity of role-playing

Model Capabilities

Role-playing dialogue generation

Context understanding

Multi-turn dialogue maintenance

Personalized response generation

Use Cases

Entertainment

Anime character interaction

Engage in realistic dialogues with anime characters like Haruhi Suzumiya

Generate dialogue responses that align with the character's personality

Creative writing

Novel character dialogue generation

Generate dialogue content for specific characters in novel writing

Assist writers in conceptualizing character interaction scenarios

🚀 ChatHaruhi Role - Playing with Qwen 7B

This project fine - tunes and tests the Qwen 7B model, enabling it to perform role - playing tasks.

🚀 Quick Start

This script fine - tunes and tests the Qwen 7B model to give Qwen 7B the capability of role playing.

The project link is Chat - Haruhi - Suzumiya.

The 118K training data was collected by LC1332.
The model was trained by goodnessSZW.
The Qwen inference code was written by hhhwmws0117 and integrated into Chatharuhi. Currently, this person is responsible for model maintenance and bug fixing.
LC1332 wrote the prompt organization function inside ChatHaruhi.

You can see a Harry Potter test at Harry_Potter_test_on_Qwen7B.ipynb.

📦 Installation

Not applicable as no specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

Load the functions:

from transformers import AutoTokenizer, AutoModel, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("silk-road/ChatHaruhi_RolePlaying_qwen_7b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("silk-road/ChatHaruhi_RolePlaying_qwen_7b", device_map="auto", trust_remote_code=True)
model = model.eval()

Refer to the specific notebook at ChatHaruhi_x_Qwen7B.ipynb.

from ChatHaruhi import ChatHaruhi

chatbot = ChatHaruhi( role_name = 'haruhi', max_len_story = 1000 )

prompt = chatbot.generate_prompt(role='阿虚', text = '我看新一年的棒球比赛要开始了！我们要去参加吗？')

response, history = model.chat(tokenizer, prompt, history=[])
print(response)

chatbot.append_response(response)

Currently, it supports multiple role format loading methods, including role_name, role_from_hf, and role_from_jsonl.

📚 Documentation

Not applicable as no detailed documentation is provided in the original document.

🔧 Technical Details

Not applicable as no technical details are provided in the original document.

📄 License

The project is licensed under the Apache - 2.0 license.

Property	Details
Model Type	Qwen 7B fine - tuned for role - playing
Training Data	- silk - road/ChatHaruhi - Expand - 118K - silk - road/ChatHaruhi_NovelWriting

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご