Nicrron is the unified gateway to every major AI model.Ninoo is the coding agent that brings it to your terminal. Same account, same credits, pick your interface. And NiNI Labs is where we research what comes next.
The unified interface for LLMs — for your applications.
The coding agent for your terminal — for developers.
The same agent in your browser — for everyone else.
Today's computing systems will undergo a massive transformation to efficiently and scalably serve AI workloads. NiNI Labs is an applied research lab dedicated to envisioning the next generation of these systems.
Serverless inference for AI agents — from simple agents to multi-agent systems.
Bring the agentic pipeline you already run. Simple single agents and multi-agent systems with custom logic are both first-class.
Connect multiple models with non-model stages like search, plus your own data sources, into one serverless pipeline.
The platform schedules, orchestrates, and optimizes every stage, so you spend your time adding capabilities to your agents.
Powering AI at supercomputing scale
Traditional data centers were built for general-purpose workloads. Nicroo Core AI data centers are built for AI.
Built for general-purpose workloads
Built for AI
The same gateway, the same key namespace, the same balance — whether you call it from code or talk to it in a terminal.
// Works with the OpenAI SDK you already use const client = new OpenAI({ baseURL: "https://api.nicrron.ai/v1", apiKey: process.env.NICRRON_API_KEY, }); const res = await client.chat.completions.create({ model: "nicrron/auto", messages, });
❯ fix the flaky retry test in billing ◆ grep(retry, tests/billing) ╰ 3 matches in tests/billing/retry.test.ts ◆ edit(tests/billing/retry.test.ts) ╰ + vi.useFakeTimers() ◆ bash(npx vitest run tests/billing) ╰ 12 passed (12) ◜ Weaving… (14s · 3.2k tokens · esc to interrupt)
Point your OpenAI SDK at Nicrron's gateway and get every provider through one API key, with routing, failover, and caching handled for you.
Start with the Nicrron APIInstall Ninoo and hand it real tasks — it reads the repo, writes the diffs, runs the tests, and asks before anything touches disk.
Install NinooThat's the point. One balance meters your production traffic and your terminal sessions together, with every request in one ledger.
Create the one accountNiNI Cloud, from NiNI Labs, is serverless inference for AI agents: import your pipeline, chain models with search and data, and let the platform scale it.
Get early access to NiNI CloudTop up once on nicrron.ai; API calls and Ninoo sessions draw from the same wallet at the same per-token rates.
A single Nicrron account covers both. Ninoo signs in through your browser and mints its own named key you can revoke anytime.
Every request — from your code or from Ninoo — appears in the same dashboard logs with its exact cost.
One signup at nicrron.ai/register covers the API and Ninoo.
Top up in the dashboard — and Nicrron matches every $100 you deposit with $100 in bonus credits, up to $500 per rolling 90 days (applied after a brief account review, typically within 24 hours).
Swap your OpenAI SDK's base URL to https://api.nicrron.ai/v1, or bring the models to your terminal:
$ curl -fsSL https://ninoo.ai/install.sh | shNo. One Nicrron account covers both. Ninoo's browser login mints an API key on that same account, and its usage appears in the same dashboard.
The CLI is free. You pay only the per-token cost of the model calls it makes, at the same published rates as calling the gateway directly — no subscription, no markup on top.
The live catalog on nicrron.ai lists 65+ models across ten providers — chat, reasoning, embeddings, image and video generation, and speech. Ninoo can switch between any chat model with /model, or use routing modes like nicrron/auto.
On nicrron.ai. Top up there, set auto top-up, and watch the shared ledger. Nicrron's deposit-match promotion — every $100 matched with $100 in bonus credits, up to $500 per rolling 90 days — applies to credits you spend in either product.
Chat prompts and completions are never stored by the gateway — from the API or from Ninoo. Generated media is kept 30 days so you can re-download it; the details are in the privacy policy on nicrron.ai.
An applied research lab dedicated to envisioning the next generation of computing systems — the ones that will efficiently and scalably serve AI workloads. NiNI Cloud, serverless inference for AI agents, is its first service.
Not yet. It is in development at NiNI Labs; email support@nicrron.ai with the subject "NiNI Cloud waitlist" to get early access. Pricing and billing details will be announced with the launch.
AI data centers built for AI. Traditional data centers were built for general-purpose workloads; Nicroo Core is built to power AI at supercomputing scale. Email support@nicrron.ai with the subject "Nicroo Core" to start a conversation.