Ghonk API/Docs

The Ghonk v4 model has been released as an experimental model, its ID is "ghonk_v4_062126", and it can be tested for free with limitations on usage, resources, and functions. It is partially available in the Ghonk app. What's new in this model?: - Improved advanced reasoning resulting in complete answers and fewer "hallucinations" - More advanced code generation with direct integration to the GPT 5.2 Codex - Fewer text generation errors and faster responses - Increased security and security filters to prevent improper use, API abuse, and other issues...

The API can only be used in the "PUT" call method via the endpoint: https://openrouter.ai/api/v1/chat/completions Example call for the basic model (Ghonk-v3) curl https://openrouter.ai/api/v1/chat/completions \ -H "Authorization: Bearer $OPENROUTER_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "@preset/ghonk-v3", "messages": [ { "role": "user", "content": "Hello! How are you today?" } ] }' Only one API can be used for Ghonk model calls, and the free public API Key is: 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 * It has a daily limit of 100,000 credits.

The Ghonk v3 family of models contains different models for different areas, including: - Ghonk v3: For casual conversation, standard speed and reduced features. - Ghonk v3 Fast: Ghonk v3 model with noise reduction, removal of heavy/medium reasoning, ideal for short conversations. - Ghonk v3 Reasoning: For complex tasks and activities such as mathematics, deep learning, etc. - Ghonk v3 Code: For developers, optimized for code generation and correction, adapted for medium reasoning.