You can choose between the following OpenAI models in your extension settings. Each model has its own balance of performance, cost, and capability.
🧠 gpt-3.5-turbo
- Speed: Very fast response time
- Cost: Lowest cost model
- Best for: Simple tasks like generating product descriptions, SEO metadata, FAQs, short summaries, etc.
- Usage Suggestion: Ideal for reducing token usage costs while still generating decent quality content.
- Approximate Cost:
- Input tokens: ~$0.0005 per 1K tokens
- Output tokens: ~$0.0015 per 1K tokens
🚀 gpt-4-turbo (Legacy Turbo Models: gpt-4-0125-preview / gpt-4-1106-preview)
- Speed: Fast, slightly slower than 3.5, but delivers much higher quality
- Cost: Affordable for high-quality output
- Best for: Complex prompts, long-form articles, accurate SEO content, multi-language generation, and structured data outputs
- Usage Suggestion: Recommended when you need higher accuracy, better grammar, and polished tone. Suitable for important content like landing pages, blog posts, and marketing copy.
- Approximate Cost:
- Input tokens: ~$0.01 per 1K tokens
- Output tokens: ~$0.03 per 1K tokens
💡 gpt-4o (Latest and Most Advanced)
- Speed: Extremely fast — faster than all previous models
- Cost: Same pricing as GPT-4 Turbo, but significantly more capable
- Best for: Everything from basic content to complex tasks, high-context conversations (up to 128K tokens), and highly accurate SEO writing
- Usage Suggestion: The best overall choice — combines performance, speed, and accuracy.
- Approximate Cost:
- Input tokens: ~$0.01 per 1K tokens
- Output tokens: ~$0.03 per 1K tokens
⚙️ How to Configure the Model in OpenCart
To select your preferred GPT model:
- Go to your OpenCart Admin Panel
- Navigate to:
SEO Extensions > AI Generator
- Locate the field labeled “GPT Model”
- Enter one of the following model names:
- For GPT 3.5:
gpt-3.5-turbo
- For GPT 4 Turbo:
gpt-4-1106-preview
(or usegpt-4-0125-preview
) - For GPT 4o (Recommended):
gpt-4o
- For GPT 3.5:
- Click Save
Notes
- The actual cost depends on the number of tokens processed per request.
- You must have an active OpenAI API key added in the extension settings.
- If the entered model name is invalid or unavailable, the API will return an error.