Skip to main content

Using Chinda LLM 4B with Ollama - Complete User Guide

🎯 Introduction​

Chinda LLM 4B is an open-source Thai language model developed by the iApp Technology team, capable of thinking and responding in Thai with high accuracy using the latest Qwen3-4B architecture.

Ollama is a powerful command-line tool that allows you to easily run large AI models locally on your computer with minimal setup and maximum performance.

🚀 Step 1: Installing Ollama​

Download and Install Ollama​

01_ollama_download_page.webp

For macOS and Linux:​

curl -fsSL https://ollama.com/install.sh | sh

For Windows:​

  1. Go to https://ollama.com/download
  2. Download the Windows installer
  3. Run the installer and follow the installation wizard

Verify Installation​

After installation, open your terminal or command prompt and run:

ollama --version

You should see the Ollama version information if the installation was successful.

$ ollama --version
ollama version is 0.9.0

🔍 Step 2: Download Chinda LLM 4B Model​

Downloading the Model​

Once Ollama is installed, you can download Chinda LLM 4B with a simple command:

ollama pull iapp/chinda-qwen3-4b

The model is approximately 2.5GB in size and will take some time to download depending on your internet speed. You'll see a progress bar showing the download status.

$ ollama pull iapp/chinda-qwen3-4b
pulling manifest
pulling f2c299c8384c: 100% ▕██████████████████▏ 2.5 GB
pulling 62fbfd9ed093: 100% ▕██████████████████▏  182 B
pulling 70a7c2ca54f5: 100% ▕██████████████████▏  159 B
pulling c79654219fbe: 100% ▕██████████████████▏   74 B
pulling e8fb2837968f: 100% ▕██████████████████▏  487 B
verifying sha256 digest
writing manifest
success

Verify Model Download​

To check if the model was downloaded successfully:

ollama list

You should see chinda-qwen3-4b:latest in the list of available models.

$ ollama list
NAME                           ID              SIZE      MODIFIED       
iapp/chinda-qwen3-4b:latest    f66773e50693    2.5 GB    35 seconds ago

⚙️ Step 3: Basic Usage​

Start Chatting with Chinda LLM​

To start an interactive chat session with Chinda LLM 4B:

ollama run iapp/chinda-qwen3-4b

This will start an interactive chat session where you can type your questions in Thai and get responses immediately.

Example Conversation​

ollama run iapp/chinda-qwen3-4b
>>> สวัสดีครับ ช่วยอธิบายเกี่ยวกับปัญญาประดิษฐ์ให้ฟังหน่อย

# Chinda LLM will respond in Thai explaining artificial intelligence

>>> /bye  # Type this to exit the chat

Single Question Mode​

If you want to ask a single question without entering chat mode:

ollama run iapp/chinda-qwen3-4b "อธิบายเกี่ยวกับปัญญาประดิษฐ์ให้ฟังหน่อย"

🌐 Step 4: API Server Usage​

Starting the Ollama API Server​

For developers who want to integrate Chinda LLM into their applications:

ollama serve

This starts the Ollama API server on http://localhost:11434

Using the API with curl​

Basic API Call:​

curl http://localhost:11434/api/generate -d '{
  "model": "iapp/chinda-qwen3-4b",
  "prompt": "สวัสดีครับ",
  "stream": false
}'

Chat API Call:​

curl http://localhost:11434/api/chat -d '{
  "model": "iapp/chinda-qwen3-4b",
  "messages": [
    {
      "role": "user",
      "content": "อธิบายเกี่ยวกับปัญญาประดิษฐ์ให้ฟังหน่อย"
    }
  ]
}'

Using with Python​

import requests
import json

def chat_with_chinda(message):
    url = "http://localhost:11434/api/generate"
    data = {
        "model": "iapp/chinda-qwen3-4b",
        "prompt": message,
        "stream": False
    }
    
    response = requests.post(url, json=data)
    return response.json()["response"]

# Example usage
response = chat_with_chinda("สวัสดีครับ")
print(response)

💬 Step 5: Advanced Usage​

Setting Custom Parameters​

You can customize the model's behavior by setting parameters:

ollama run iapp/chinda-qwen3-4b
>>> /set parameter temperature 0.7
>>> /set parameter top_p 0.9
>>> /set parameter top_k 40

Model Information​

To see detailed information about the model:

ollama show iapp/chinda-qwen3-4b

Usage Examples​

Let's test with different types of questions:

Mathematics Question:​

ollama run iapp/chinda-qwen3-4b "ช่วยแก้สมการ 2x + 5 = 15 ให้หน่อย"

Document Writing:​

ollama run iapp/chinda-qwen3-4b "ช่วยเขียนจดหมายขอบคุณลูกค้าให้หน่อย"

Programming Help:​

ollama run iapp/chinda-qwen3-4b "เขียนฟังก์ชัน Python สำหรับหาเลขคู่ให้หน่อย"

🔧 Troubleshooting​

Common Issues and Solutions​

Issue: Model Not Found​

Error: pull model manifest: file does not exist

Solution: Make sure you typed the model name correctly:

ollama pull iapp/chinda-qwen3-4b

Issue: Ollama Service Not Running​

Error: could not connect to ollama app, is it running?

Solution: Start the Ollama service:

ollama serve

Issue: Out of Memory​

If you encounter memory issues, you can try:

ollama run iapp/chinda-qwen3-4b --low-memory

Performance Tips​

  1. Close unnecessary applications to free up memory
  2. Use SSD storage for better model loading speed
  3. Ensure sufficient RAM (recommended: 8GB or more)

🎯 Suitable Use Cases​

✅ What Chinda LLM 4B Does Well​

  1. Document Drafting - Help write letters, articles, or various documents in Thai
  2. RAG (Retrieval-Augmented Generation) - Answer questions from provided documents
  3. Mathematics Questions - Solve math problems of various levels
  4. Programming - Help write code and explain functionality in Thai
  5. Language Translation - Translate between Thai and English
  6. Creative Writing - Generate stories, poems, or creative content
  7. Educational Content - Explain concepts and provide learning materials

❌ Limitations to Be Aware Of​

Don't ask for facts without context such as:

  • Latest news events
  • Specific statistical data
  • Information about specific people or organizations
  • Real-time information

Since Chinda LLM 4B is a 4B parameter model, it may generate incorrect information (hallucination) when asked about specific facts.

🚀 Model Specifications​

  • Size: 2.5GB (quantized)
  • Context Window: 40K tokens
  • Architecture: Based on Qwen3-4B, optimized for Thai language
  • Performance: Fast inference on consumer hardware
  • Memory Requirements: Minimum 4GB RAM, recommended 8GB+

🔮 What's Coming​

The iApp Technology team is developing a new, larger model that will be able to answer factual questions more accurately. It is expected to be released soon.

📚 Additional Resources​

Quick Reference Commands​

# Install Ollama
curl -fsSL https://ollama.com/install.sh | sh

# Download Chinda LLM 4B
ollama pull iapp/chinda-qwen3-4b

# Start chatting
ollama run iapp/chinda-qwen3-4b

# List installed models
ollama list

# Remove a model
ollama rm iapp/chinda-qwen3-4b

# Start API server
ollama serve

🎉 Summary​

Chinda LLM 4B with Ollama is an excellent choice for those who want to use Thai AI models locally with maximum performance and minimal setup. The command-line interface provides powerful features for both casual users and developers.

Key advantages of using Ollama:

  • Fast performance with optimized inference
  • Easy installation with simple commands
  • API support for integration into applications
  • No internet required after initial download
  • Privacy-focused - all processing happens locally

Start using Chinda LLM 4B today and experience the power of Thai AI running locally on your machine!

Built with ❤️ by iApp Technology Team - For Thai AI Development