Turn - Detection - Cocalai - vllm Open - Source Model - Precise detection of conversation turns, with an accuracy rate as high as 96.22%

Turn Detection Cocalai Vllm

Developed by anonymguy

This model is a dialogue turn detection model fine-tuned based on Qwen3-0.6B. It uses the Unsloth and Huggingface TRL libraries to accelerate training, achieving an accuracy of 96.22% in the dialogue turn detection task.

Large Language Model

Transformers

EnglishOpen Source License:Apache-2.0 #Dialogue turn detection #Efficient fine-tuning #Customer support optimization

Downloads 351

Release Time : 6/15/2025

Model Overview

A language model optimized specifically for the dialogue turn detection task, capable of accurately identifying turn transition points in dialogues.

Model Features

Training acceleration

Achieve a 2x increase in training speed using the Unsloth and Huggingface TRL libraries

Multi-dataset training

Train by combining the latishab/turns-2k dataset and a synthetic dataset of 140 samples

High accuracy

Achieve an accuracy of 96.22% in the dialogue turn detection task

Model Capabilities

Dialogue turn detection

Text generation

Use Cases

Customer support

Dialogue analysis

Automatically detect the dialogue turns between customers and support staff

Accurately identify 96.22% of dialogue turn transitions

Dialogue system

Chatbot

Help chatbots more accurately determine when to respond

Property	Details
Base Model	unsloth/Qwen3-0.6B-Base
Tags	text-generation-inference, transformers, unsloth, qwen3, turn-detection
License	apache-2.0
Language	en
Datasets	latishab/turns-2k

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

Turn Detection Cocalai Vllm

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 Uploaded finetuned model

🚀 Quick Start

Model Information

Model Details

Visualization

Acknowledgment

📄 License