Step 2

Sign Up

After downloading the IDE, go to the signup page and create your account.

Step 3

Sign In

Open the IDE and click the profile icon at the top-right to sign in.

Step 4

Disable Auto Mode

After creating your account, come back to the IDE. At the bottom of the chat panel, click the Auto button and turn off Auto Mode.

Step 5

Add a Custom Model

Auto Mode disable karne ke baad, Add Model button par click karein. Modal khulne par Custom Config tab select karke neeche di gayi details bharein.

Custom Config tab mein jaakar Base URL, Model ID aur API Key daalein.

Agar model add karne par "this is not multimodal" error aaye, to Multimodal option ko disable kar dein. Baaki settings default rakhein.

Kuch bhi rakh sakte hain.

Input = 184000, Output = 32000 set karein (recommended).

Default 200 ki jagah 430 likhein.

Sab kuch bharne ke baad neeche Add Model button par click karein.

Custom Config tab mein jaakar Base URL, Model ID aur API Key daalein.

Agar model add karne par "this is not multimodal" error aaye, to Multimodal option ko disable kar dein. Baaki settings default rakhein.

Kuch bhi rakh sakte hain.

Input = 184000, Output = 32000 set karein (recommended).

Default 200 ki jagah 430 likhein.

Sab kuch bharne ke baad neeche Add Model button par click karein.

Step 6

Get API Credentials

Kisi bhi AI API provider platform (jaise NVIDIA, Groq, OpenRouter, etc.) se API Key, Base URL, aur Model ID kaise lein.

Apne desired AI API provider (jaise NVIDIA, Groq, OpenRouter, etc.) ki website par jaakar account banayein ya sign in karein.

Dashboard mein jaakar API Keys, Developer, ya Settings section dhundhein.

Models ki Pricing page par jaakar check karein. Jin models ka input/output cost $0 ya $0.00 ho, wo bilkul free hain. Free models select karein taaki aap bina paise kharch kiye use kar saken.

Nayi API key generate karein, use descriptive naam dein, aur key ko copy karlein (ek baar hi dikhti hai!).

Platform ka API endpoint/Base URL note karein. Example: https://integrate.api.nvidia.com/v1 ya https://api.groq.com/openai/v1

Available models ki list mein se apna desired model select karein aur uska exact Model ID copy karein (e.g., meta/llama-3.1-405b-instruct).

Teeno details (Base URL, Model ID, API Key) Step 5 mein jakar bharein aur Add Model button dabayein.

💡

Pro Tip: Har platform ka interface thoda different ho sakta hai, lekin basic process same rehta hai. Provider ki documentation zaroor check karein.

Step 7

NVIDIA Models Use Karein

Note: NVIDIA par kafi load rehta hai, isliye pehle kam load wale models try karein jaise stepfun-ai/step-3.7-flash, qwen/qwen3.5-397b-a17b, qwen/qwen3-coder-480b-a35b-instruct, ya mistralai/mistral-large-3-675b-instruct-2512.

build.nvidia.ai par jayen aur Models explore karein.

Scroll to view next

Model card par click karein aur Prototype section se Base URL aur Model ID copy karein.

Scroll to view next

Top right corner mein profile icon par click karein, phir API Keys par click karke API key hasil karein.