Onyx
Configure Onyx with AICredits as an OpenAI-compatible model provider for enterprise search, chat, agents, and RAG.
Use this page with an AI assistant
Opens a new chat with this docs URL and the correct AICredits base URLs.
Onyx is an open-source AI chat and enterprise search platform with RAG, agents, permissions, analytics, and admin controls. Add AICredits as an OpenAI-compatible LLM provider to centralize model access and INR billing.
AI Model Settings
In the Onyx admin model settings, add a custom OpenAI-compatible provider:
| Setting | Value |
|---|---|
| Provider name | AICredits |
| Base URL | https://api.aicredits.in/v1 |
| API key | sk-your-aicredits-key |
| Chat model | Pick from Models |
Recommended chat models:
openai/gpt-4o-mini
anthropic/claude-sonnet-4.5
google/gemini-2.5-flashEmbeddings
For document search and retrieval, configure embeddings separately when your Onyx deployment exposes that setting:
Base URL: https://api.aicredits.in/v1
API key: sk-your-aicredits-key
Model: text-embedding-3-smallUse a dedicated AICredits key for Onyx. Enterprise search tools can generate both ingestion-time embedding usage and chat-time LLM usage.
Verify
Run a short chat test before connecting large document sources:
Say "AICredits via Onyx" and nothing else.Then check Dashboard -> Usage.
Troubleshooting
Chat works but search quality is poor - Confirm embeddings are configured and documents have been re-indexed.
401 Unauthorized - Replace the model provider key with a current AICredits key.
Unexpected usage - Separate ingestion keys from chat keys if you want clearer budget controls.