Scattered integrations
- Different endpoints for different models
- Provider-specific setup in your app
- Extra work when your needs change
Connect to GPT, Gemini, Claude, and more through one familiar API. Choose a model, or let Kora find a route for you.
Live on Kora
One integration gives your application access to the models available on Kora.
Build with a familiar interface and keep the model choice flexible.
Use the OpenAI client with Kora’s base URL and your API key.
Keep the integration simple.Ask for a specific model, or choose Auto and let Kora pick an available route.
Change models without changing apps.Receive text as it arrives in chat or through the API.
See the response take shape.Review the model used, response time, and token usage in your workspace.
Know what each request did.Go from an idea to a working request without rebuilding your stack.
Point an OpenAI-compatible client at Kora. Use one API key and one base URL to reach the models on your plan.
Try prompts in browser chat, switch models, and keep your conversations together as you explore an idea.
See current model access in one catalogue and choose the route that fits the work in front of you.
Connect once, choose a route, and keep building.
Use your Kora key and the OpenAI-compatible base URL.
Pick a model by name or let Auto choose an available route.
Stream the answer and review usage in your workspace.
client = OpenAI(
base_url="https://koranet.xyz/v1",
api_key="kora-q-..."
)one base URL / one keyStreamed back through the same familiar API.
Real traffic and current model access from the running service.
Browse models, test an idea in chat, then use the same route from your app.
One place to start. More ways to build.Open Chatinput + output tokens served all time
models currently available across plans
OpenAI-compatible requests, Python clients, and your own applications.
Start in chat, or send a request from your own code.
Connect with the OpenAI client and send a chat completion to Kora.
Python / API / Modelsfrom openai import OpenAI
client = OpenAI(
base_url="https://koranet.xyz/v1",
api_key="kora-q-..."
)
reply = client.chat.completions.create(
model="auto",
messages=[{"role": "user", "content": "Hello, Kora."}]
)Explore the models your plan can reach, and switch routes as your work changes.
Catalogue / Free / ProStart free. See the available models before you choose.
For getting started
For heavier use
A few quick answers before you start building.
Kora is a model router with an OpenAI-compatible API and browser chat. It brings multiple available AI models behind one base URL.
If your app uses the OpenAI client, set Kora as the base URL and use a Kora API key. Check the model catalogue for the models your plan can access.
Yes. Sign in and open Chat to try prompts with the models available on your plan.
Yes. Choose an available model by its public ID, or use Auto when you want Kora to select a route.
Your dashboard shows request history and token usage. The live totals above reflect traffic served by Kora.
Availability can change. The live model catalogue shows what Free and Pro can currently access.
One account, one familiar API, and a whole catalogue to explore.