New miracache hardware partner programme 

MIRASTACK LABS

AI capacity, sold by time. Not by the token.

We host and run open-weight models on our own GPUs, and sell that capacity the way broadband is sold: you pay for the window, never per unit. From a private AI factory in your data centre to one developer's day pass.

Now · miracache

Give your DDR4 servers a second life in AI inference.

miracache turns the DRAM of older servers into a shared memory tier for LLM inference, reached over RDMA. It is built to plug into vLLM and SGLang, so they can keep conversation state instead of computing it again. We are starting the first hardware tests now. Bring your hardware and run them with us.

02 — Offerings

Four ways we work with you

  • IaaS · Sovereign and regulated

    AI Factory

    An AI factory in your own data centre, on your hardware or ours, on your domain. We design it, bring it live and run it. Nothing leaves your walls.

    Explore the AI Factory
  • PaaS · Model engineering and serving

    infinitetokens

    Infinite tokens for a finite time. An AI platform for scoped model engineering and serving: training clusters and inference reserved for your work on your own endpoint, <company>.infinitetokens.ai, for 3, 6, 9 or 12 months. Billed for the platform and the term, never per token.

    Talk to us
  • SaaS · Teams

    unlimitedtokens

    Open models for your whole workforce, billed per seat by time, never per token. For organisations from 60 seats, on quarterly or yearly agreements.

    Talk to us Flat-rate LLM API for teams
  • Individuals

    endlesstokens

    A flat time pass with unlimited tokens on open models, for developers and their coding agents. Self-serve, in minutes.

    Visit endlesstokens.ai

03 — Why time, not tokens

Stop rationing your thinking.

Metered APIs bill every token, so every agent loop and every experiment becomes a cost decision. We sell capacity by time: the cost is agreed before you start, and your people and systems stop rationing their thinking.

  • Predictable costAgreed up front, per seat or per term.
  • No token meteringUse the capacity you bought.
  • Terms that fitSeats, sessions, keys and terms are set per agreement.

Unlimited tokens or per-token pricing: how to choose

04 — Our technical moat

The full path, from GPU to API.

Every model we offer runs on GPUs we host and operate. We do not proxy your requests to another provider and we do not resell anyone else's capacity. Frontier proprietary models stay with their labs.

  • Hosted, not resoldOur own GPUs and our own gateway, with no hidden third party in the path.
  • Measured before it is soldWe replay real agent sessions and size capacity per GPU before we put it on sale.
  • Researched in the openWe work with an independent inference lab that publishes its results, one claim at a time.

05 — Who we build for

Built for people who run AI for a living.

  • EnterprisesBringing AI to every desk.
  • Model and research teamsTraining, tuning and evaluating at scale.
  • Regulated financial institutionsAI that stays inside your audit and data boundary.
  • Defence and public sectorAir-gapped AI, run on your own ground. On-premises sovereign AI
  • Critical infrastructure operatorsHuman-approved AI for systems that cannot stop.
  • Sovereign data centre operatorsAI factories built, run and kept in-country. On-premises sovereign AI
  • Developers and AI-native start-upsA flat time pass, never a token bill.

Talk to us

Tell us what you need to run.

Every agreement is shaped to the workload: seats, sessions, capacity, term and location.