Dify
Connect Dify to AICredits with an OpenAI-compatible model provider for chatbots, RAG apps, agents, and workflows.
Use this page with an AI assistant
Opens a new chat with this docs URL and the correct AICredits base URLs.
Dify is a visual platform for building LLM applications, agents, and RAG workflows. Add AICredits as an OpenAI-compatible model provider to use Claude, Gemini, GPT, DeepSeek, and other models from one INR-billed API key.
Setup
In Dify, add an OpenAI-compatible model provider:
| Setting | Value |
|---|---|
| Provider name | AICredits |
| API endpoint URL | https://api.aicredits.in/v1 |
| API key | sk-your-aicredits-key |
| Model type | Chat |
| Model name | Pick from Models |
Use exact AICredits model IDs such as:
openai/gpt-4o-mini
anthropic/claude-sonnet-4.5
google/gemini-2.5-flash
deepseek/deepseek-v4-flashKeep the /v1 suffix in the endpoint URL. Dify sends OpenAI-compatible requests under that base path.
Use In Apps
After saving the provider:
- Open your Dify app.
- Go to the model settings for the app, workflow, or agent node.
- Choose
AICreditsas the provider. - Select or type the model ID.
- Run one low-cost test prompt.
For RAG apps, start with a fast chat model such as openai/gpt-4o-mini for retrieval-grounded answers. Use a larger reasoning model only for final responses that need higher quality.
Hostinger Deployment Notes
If you deploy Dify as a Hostinger container, store the AICredits key in Dify's provider settings, not in prompts or workflow variables. Use a dedicated AICredits API key with a monthly budget cap for each shared Dify workspace.
Verify
Create a simple chat app and ask:
Reply with "AICredits via Dify" and nothing else.Then check Dashboard → Usage to confirm the request appears with the expected model and cost.
Troubleshooting
401 Unauthorized — Replace the provider key with a current AICredits API key.
Model not found — Use the exact model ID from Models.
Connection fails — Confirm the endpoint is https://api.aicredits.in/v1, not the bare domain.