What does pre-launch actually mean?
It means you get early access, and we get your feedback.
Try it, break it, tell us what to improve. We’re building this with you.
Create your API key now: https://t.co/3TLimOCvuM
PSA if you're building on Gemini 4 Argon: the launch price is introductory.
$2 in / $10 out per million tokens until Nov 30. From Dec 1 it doubles to $4 / $20.
Planning your unit economics on intro pricing is like budgeting your salary around the Diwali bonus.
Indian dad reading the card statement:
OPENAI : ₹1,742
ANTHROPIC : ₹2,310
GOOGLE : ₹980
FOREX MARKUP : ₹151
"Beta, ye sab kaun log hain? Aur dollar mein kyun?"
One key. One UPI top-up. One GST bill. Papa approved
Pre-launch is live: https://t.co/g6JiAQutyj
A developer in India should not need a dollar card and twenty accounts to use AI.
We are building the fix. It is called Rapid Router.
Pre-launch is live: https://t.co/3TLimOCvuM
Google just shipped Gemini API CLI (preview): agents generate images, video, speech and music from the terminal. No more guessing model strings.
Now fix the other guess: which key, which card. One Rapid Router key, 600+ models, INR via UPI. https://t.co/cXEGKOkhfD
UPI changed how India pays. Rapid Router is changing how India pays for AI.
One AI API, one key, many models, paid in rupees.
Pre-launch is live: https://t.co/cXEGKOkP5b
When one provider goes down, an app built on it goes down too. Rapid Router will move the request to another route that serves the same model.
Pre-launch is live. https://t.co/3TLimOCvuM
@rsumit123@AbhiBrooo_ Fair question. Same idea, built around how India pays: top up in rupees over UPI, get a GST invoice, and FEMA and TDS are handled for you. No dollar card, no forex markup, and models at provider list price.
@igrageop Thanks for the honest take. Per token you pay provider list price, so it should match going direct. 10 tok/s on a flash model isn't right though. Which model, and roughly when? DM us and we'll dig in today.
@test_worldend Fair, cost is what matters. Every model is charged at provider list price, same as going direct. The only extra is a small fee once per top-up, not per call, and there's no forex markup from your card. Breakdown: https://t.co/8Qemxkp2FW
@rarexraa@grok Claude Max is great for chat and coding tools. But the moment you build a product that calls the API, it's pay-per-token on a separate account, usually with a dollar card. That's the part we fix: UPI top-up, one key, GST invoice.
Gemini 4 Argon built an interactive 3D anatomy model from three open datasets. Biology class would've hit different 🫀
Argon is already in the Rapid Router catalog, ready for when we open the doors.
Gemini 4 Argon combined three open-source datasets to build an interactive 3D visualization of human anatomy.
Toggle different layers in motion, isolate muscles, or zoom inside a beating heart.
OpenAI just made prompt caching last 12 hours in the API. If your agent reuses the same instructions and tools all day, that's a real cut in input costs.
Building Rapid Router so Indian devs get one key for models like these, paid in INR via UPI. Pre-launch.
You can now choose 12-hour prompt caching in the API.
Persistent agents, like dots, reuse context throughout the day.
You can keep shared instructions, tools, and context cached longer and pay the lower cached-input rate, even when requests are hours apart.
@Nickonlyfru@nsitharaman Agreed, that would be the dream 🙌 Until then, we'll handle the dollar side for you: top up in rupees with UPI, get a GST invoice, call any of 640+ models. Helping New India build, the new way
@Itsmenxyn all three are already on the pre-launch 😎 anthropic/claude-opus-5-5, google/gemini-4-argon and openai/gpt-6.1-sol, at provider list price, on one key. Come take them for a spin: https://t.co/Al37ouGnWv
Bangalore people, you want one dosa, not a 5-course tasting menu. You want one AI API key, not 5 provider accounts, 5 dashboards and 5 bills.
One account, one key, many models.
Pay with UPI. Get a proper GST bill.
Rapid Router.
Pre-launch is live: https://t.co/cXEGKOkP5b