Tool integrations

Because the API is OpenAI-compatible, most tools just need two settings: the base URL and your key. Below are copy-paste configs for popular ones.

Base URL: https://api.studio.whalli.com/v1 · API key: a wh_live_… key from the Console.

OpenAI SDKs#

Point the client at the base URL — nothing else changes.

python

from openai import OpenAI

client = OpenAI(base_url="https://api.studio.whalli.com/v1", api_key="wh_live_...")

javascript

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.studio.whalli.com/v1", apiKey: process.env.WHALLI_API_KEY });

OpenCode#

Add Whalli as an OpenAI-compatible provider in ~/.config/opencode/opencode.json:

json

{
  "provider": {
    "whalli": {
      "npm": "@ai-sdk/openai-compatible",
      "options": {
        "baseURL": "https://api.studio.whalli.com/v1",
        "apiKey": "{env:WHALLI_API_KEY}"
      },
      "models": {
        "gpt-5.5": { "name": "Whalli · gpt-5.5" }
      }
    }
  }
}

Cline / Cursor#

In the model settings, choose OpenAI Compatible and set:

text

Base URL:  https://api.studio.whalli.com/v1
API Key:   wh_live_...
Model:     gpt-5.5   (any id from GET /v1/models)

Ollama-compatible clients#

Any app that targets an OpenAI/Ollama-style endpoint works. Set its OpenAI base URL and key:

bash

OPENAI_BASE_URL=https://api.studio.whalli.com/v1
OPENAI_API_KEY=wh_live_...

LiteLLM#

yaml

model_list:
  - model_name: whalli-gpt-5.5
    litellm_params:
      model: openai/gpt-5.5
      api_base: https://api.studio.whalli.com/v1
      api_key: os.environ/WHALLI_API_KEY

Don't see your tool? If it supports an "OpenAI compatible" or custom base URL, it works — set the base URL to https://api.studio.whalli.com/v1 and use your key.