One API for many models · Billed in rupees
We test your real workload across many models and show which calls can run for less. Then one API routes each call to the best-value model. Billed in rupees with GST invoices.
One API for many AI models. IndicStack routes each call to the best-value model that meets your rules, proven on your own workload. Billed in rupees with GST invoices.
Most teams pay top-model prices for every call. We show which of yours can run on a lower-cost model, with proof on your own workload, and route them through one API. Billed in rupees.
Free for 15 days on Qwen3.8-27B. No card needed.
Switch in two lines
Point your base URL at IndicStack and use your IndicStack key. Your prompts, tools and streaming code stay as they are.
Read the quickstart →from openai import OpenAI
import os
client = OpenAI(
base_url="https://api.indicstack.ai/v1",
api_key=os.environ["INDICSTACK_API_KEY"],
)
response = client.chat.completions.create(
model="qwen38-27b",
messages=[{"role": "user", "content": "Explain what an API is."}],
)
print(response.choices[0].message.content)
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.indicstack.ai/v1",
apiKey: process.env.INDICSTACK_API_KEY,
});
const response = await client.chat.completions.create({
model: "qwen38-27b",
messages: [{ role: "user", content: "Explain what an API is." }],
});
console.log(response.choices[0].message.content);
curl https://api.indicstack.ai/v1/chat/completions \
-H "Authorization: Bearer $INDICSTACK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "qwen38-27b",
"messages": [{"role": "user", "content": "Explain what an API is."}]
}'
Developers & startups
Start free on Qwen3.8-27B for 15 days, then pay per token or choose Pro or Team.
client = OpenAI(
base_url="https://api.indicstack.ai/v1",
api_key=os.environ["INDICSTACK_API_KEY"]
)Agencies
Give each client a key. Track spend by key and bill from actual usage.
The model catalog
Compare capabilities and hosting before you build.
Hosting controls
Browse models by capability and check the hosting information.
Qwen3.8-27B runs on GPUs in India.
Check the data-flow overview and request compliance documents for your workload.
Building for a regulated client? See how we handle data →
Trust & procurement
India‑hosted models run on GPUs in Indian data centres.
Prompts and completions on India‑hosted models are never used to train models.
Model, tokens, time and key for every call.
100% Indian-owned LLP, Bengaluru.
Rupee billing your finance team can file.
Sub-processor list, data-flow overview and security questionnaire answers.
Start with the integration, hosting and billing details that affect your build.
Discuss your requirementsSign up in the console, create a key, and configure your server's OpenAI SDK with https://api.indicstack.ai/v1 and your IndicStack key.
Free Trial: free for 15 days on Qwen3.8-27B, 1 request at a time, 20 requests per minute and 1 API key. Other models need a prepaid wallet. Pro: ₹2,499/month + GST for unlimited Qwen3.8-27B and ₹2,000 credit a month for every other model, 2 requests at once, 60 requests per minute and 1 API key. Team: ₹9,999/month + GST for unlimited Qwen3.8-27B and ₹8,000 credit a month for every other model, 8 requests at once, 200 requests per minute and up to 10 API keys. Unlimited applies to Qwen3.8-27B, which we host ourselves. It runs within your plan's request limits; when those are reached, further requests wait their turn. On Qwen3.8-27B, pay ₹21 per million input tokens and ₹190 per million output tokens, before 18% GST.
New accounts can use Qwen3.8-27B free for 15 days. Other models need a prepaid wallet, with a minimum top-up of ₹500. Credit you buy never expires.
India‑hosted models run on GPUs in India. Global models run outside India on partner infrastructure; check the hosting label for the route you select.
Pay-as-you-go calls stop when your prepaid wallet reaches zero. Top up before making further pay-as-you-go calls. Pro and Team still include unlimited Qwen3.8-27B.
Create a key per client and track spend by key on the dashboard. IndicStack bills you in rupees with GST invoices.
Prompts and completions on India‑hosted models are never used to train models.
Dedicated GPU deployments in India, specialist models, custom agents, RAG and AI consulting are available through our team.
Free for 15 days on Qwen3.8-27B. No card needed.
We use essential cookies to run this site, analytics to understand how it is used, and advertising technologies to measure campaigns. You can turn off analytics and advertising with Reject all or Settings. Cookie Policy