Cheap AI Inference in India.
Multiple Models, One Single API.
Run the world's best open models — DeepSeek, GLM, Kimi, Grok, and Qwen — through a single OpenAI-compatible endpoint. Stop juggling multiple subscriptions and foreign card fees. Pay in Indian Rupees (₹) with instant UPI recharge.
Instant free welcome balance on signup — no credit card required. Transparent USD token pricing with instant UPI top-ups anytime.
curl -fsSL https://api.interference.dev/v1/models -H "Authorization: Bearer sk_live_..."|
Why use Ointerference?
Everything you need to build AI apps and agents in India without high costs, foreign transaction hassles, or API fragmentation.
- [*]70+ Models, 1 InterfaceAccess DeepSeek, GLM, Kimi, Grok, Qwen, and Llama from a single API key. Switch models with one line of code.
- [*]Cheapest AI Inference in IndiaUp to 80% lower cost than proprietary APIs. Transparent token pricing in USD ($) with instant UPI & card top-ups in ₹.
- [*]Instant UPI & Rupee (₹) PaymentsRecharge in seconds with Google Pay, PhonePe, Paytm, or Netbanking. No foreign currency markups or blocked international cards.
- [*]Drop-in OpenAI SDK CompatibilityZero code rewrites. Just change the baseURL to Ointerference in your existing OpenAI Python or TypeScript SDK.
- [*]Streaming & Tool Calling SupportNative support for real-time token streaming and function calling. Your application code executes locally on your terms.
- [*]Works with Any FrameworkPlug directly into Cursor, Vercel AI SDK, LangChain, Claude Code, Continue.dev, AutoGen, and custom agents.
How Ointerference works
Integrate in 60 seconds with 3 simple steps.
baseURL: "https://api.interference.dev/v1" in any standard OpenAI SDK. Switch between 70+ models with one parameter.Fast, Scalable, and India-Native
With < 15ms routing overhead, fair per-token billing in USD ($), and 70+ models ready on demand, Ointerference gives you dependable AI compute at fraction of standard prices.
Zero SDK Modifications
Works with your existing code in TypeScript, Python, Go, or cURL.
import { createOpenAICompatible } from '@ai-sdk/openai-compatible';
import { streamText, tool } from 'ai';
import { z } from 'zod';
const ointerference = createOpenAICompatible({
name: 'ointerference',
baseURL: 'https://api.interference.dev/v1',
apiKey: process.env.INTERFERENCE_API_KEY, // sk_live_...
});
const result = await streamText({
model: ointerference('glm-5.3'), // switch to deepseek-v4-flash, kimi-k3, grok-4.6, etc.
prompt: 'Generate market insights and check account credits.',
tools: {
getLedgerBalance: tool({
description: 'Fetch user balance in USD ($)',
parameters: z.object({ userId: z.string() }),
execute: async ({ userId }) => ({ balanceUsd: 10.00, inrEquivalent: 850 }),
}),
},
});
for await (const delta of result.textStream) {
process.stdout.write(delta);
}Transparent Token Pricing
Industry standard USD pricing per 1M tokens • Instant UPI & card top-ups • Real-time D1 ledger tracking
Developer Referral & Revenue Share
Invite builders, teams, and autonomous agent creators. Earn a continuous 20% revenue share on their inference spend for 90 days, credited directly to your balance in real time.
Of all eligible token inference spend
Per referee from qualifying deposit
Max reward per referred account
Instant bonus on first $5+ top-up
Generate your personalized referral link or custom vanity slug (e.g. /join?ref=yourname) directly in the console.
When your referees deposit $5.00 or more, they receive an instant $5.00 credit match to test models immediately.
As their AI agents stream completions and tool-calls, 20% of their compute cost is atomically credited to your balance with zero delay.
[INFERENCE_SETTLE] req_9f82ab • referee: user_b74e • model: glm-5.3 (14.2k tok)
├── referee_balance: -$0.0141 micro_usd_settled
└── referral_reward: +$0.0028 (20%) -> credited to referrer_balance (D1 TX: tx_r8190, status: settled)FAQ
Be the first to know when we release new models & gateway features
Get notified for high-throughput releases, price drops, and agent frameworks.