Limited-period launch offer
₹10,000 free credits on
Shakti Studio
India’s largest AI development and inferencing platform
Build, fine-tune and deploy production AI on serverless H100
and
L40S GPUs — with OpenAI-compatible APIs, per-minute billing
in
INR, and your data never leaving India.
Trusted by India’s AI Pioneers
Limited-period launch offer
See What Your Free Credits Can Power.
Your credits are applied against standard pay-as-you-go pricing — so you can model your real production cost during the trial.
2.5B Tokens
on compact models like Gemma 3 4B
7 Days
of 1× H100 (80GB) compute
15 Days
of 1× L40S (48GB) compute
Engineered for Excellence
Everything you need to ship
in one platform
Serverless GPUs
H100 and L40S on demand, billed per minute. Scale to zero when idle — no reservations, no lock-in.
AI Endpoints
OpenAI-compatible APIs for a growing catalogue of open-source models. Pay per token, from ₹4 per million.
Fine-tuning + LoRA
Full fine-tuning and LoRA adapters on your own data. Bring a custom model or adapt one from the catalogue.
India-hosted, sovereign
Data residency in India by design. Built for BFSI, healthcare and GCC compliance needs.
Limited-period launch offer
Transparent pricing.
No lock-in.
Pay only for what you consume. Serverless GPUs are billed per minute; AI Endpoints are billed per token. Your ₹10,000 credits draw against these same rates.
Serverless GPU — spot, per minute
Billed in INR
AI Endpoints — per million tokens
-
Compact models (e.g. Gemma 3 4B)
from ₹4
-
Frontier reasoning (e.g. DeepSeek R1)
₹484
Per-model rates for the full catalogue are on the pricing page.
Open source models
Deploy leading
open-source models. Instantly.
LLMs
16 models
DeepSeek R1 (INT4)
DeepSeek V3 (INT4)
Devstral Small (v2505)
Gemma 3 1B
Gemma 3 4B
Gemma 3 27B Instruct
Llama 3.1 8B
Llama 3.3 70B Instruct
Llama 3.3 70B Instruct
OpenAI GPT OSS 120B
Qwen 2.5 VL 72B Instruct
Qwen 3 14B
Qwen-QwQ 32B
Qwen3 VL 235B A22B Instruct
Qwen3 VL 30B A3B Instruct
Speech To Text
3 models
Whisper Large V2
Whisper Large V3
Whisper Large V3 Turbo
Image Generation
3 models
Flux Dev
Flux.1 Kontext
FLUX.2 Dev
Text To Speech
2 models
Orpheus TTS
Veena TTS
OCR
1 model
Qwen 2.5 VL 72B Instruct
Quick Start Guide
From signup to first token in three steps
Create your account, claim your free credits, connect using your API key, and start building with enterprise-grade AI models on Shakti Cloud.
1
Register and your ₹10,000 credits are applied to your account.
2
Create an API key from the Shakti Studio console.
3
Point your existing OpenAI client at our base URL. Your code just works.
Benchmarking
Shakti Studio vs Global AI platforms
For an Indian enterprise, the choice isn't raw capability
— it's where your data sits and who you can call.
What matters
- Data residency
- Compliance & governance
- Billing currency
- Serverless GPUs
- OpenAI-compatible APIs
- Fine-tuning
- Enterprise support
- Latency for India users
Global AI platforms
- US / EU regions
- Global standards
- USD, FX exposure
- Yes — but added service costs (Lambda-style invocation and platform fees)
- Yes — often with added platform and invocation charges (Lambda-style)
- Varies
- Ticket-based, offshore
- Cross-region hops
Shakti Studio
- 100% India-hosted, sovereign
- Global Standards and Build for Indian BFSI, healthcare, GCC needs
- INR, no currency risk
- Yes — per-minute billing, no added service fees
- Yes — included, no extra charges
- Full fine-tuning + LoRA
- Local, enterprise-grade
- In-country, low latency
Industries
Built for teams that can't
compromise on where AI runs
AI startups
Ship faster without burning runway on reserved GPUs. Per-minute billing, scale to zero.
GCCs
In-country capability centres running AI for global parents — with local support and INR billing.
BFSI & healthcare
Regulated-sector workloads with India data residency and sovereign hosting by design.
IT & manufacturing
Fine-tune models on proprietary data without it ever leaving Indian infrastructure.
Build AI Instantly
How to claim your credits
- step by step
The process is simple and fully transparent. From signup to
credits in your account, here's exactly what happens.
Create your account
Go to shaktistudio.ai, click Sign Up and create your account with a business email.
Submit the offer form
Fill in the credit claim form on this page so we can link the offer to your account.
Complete your KYC
Get your KYC done from within your account — a quick, one-time verification of your business details.
We call you
Our representative calls to confirm your details and answer any questions about the offer.
Credits added
Once your account is verified by us, the ₹10,000 credits are added and you're ready to build.
Apply Now
Claim your
₹10,000 credits
Four fields, two minutes. We'll ask about your workload after you're in — not before. Credits are valid for 30 days from activation.
Prefer a conversation? Talk to our team — enterprise buyers often start there.
Frequently asked questions
This offer is applicable only to net new acquisitions or new orders.
30 days from account activation. Activation happens when you verify your email, not when you submit the form — so your clock starts when you're ready to build.
Anything on Shakti Studio: serverless GPU minutes (H100, L40S), AI Endpoint tokens, and fine-tuning jobs. Credits draw against the standard pay-as-you-go rates shown above.
Nothing automatic. You'll see your balance in the console, and you choose whether to add a payment method and continue at the same pay-as-you-go rates. No auto-charges, no lock-in.
Entirely in India, in Yotta's sovereign data centres. Data residency, processing and support all stay in-country — built for BFSI, healthcare and GCC compliance requirements.
Yes. Deploy custom or fine-tuned models on serverless GPUs, or fine-tune catalogue models with full fine-tuning or LoRA and serve them behind the same OpenAI-compatible API.
No. Register with a business email and your ₹10,000 credits are applied on activation. Payment details are only needed if you continue after the credits.