The Non-Deterministic
Entropy Layer.
PureRand turns multiple independent quantum hardware sources – based on vacuum fluctuations and photon shot noise – into a continuous stream of cryptographically secure, non-deterministic randomness. We provide the missing, auditable entropy layer for cryptography, bias-sensitive AI, and scientific, medical and financial platforms that require more than “good enough” pseudo-random seeds.
- Multi-source hardware entropy from well-characterised quantum processes
- Every output ships with verifiable metadata and a hash chain (auditability)
- Stateless design: integrate via simple, secure HTTP calls
- Designed for cryptography, bias-aware AI, research, finance and gaming
Free tier: 5 API calls / month — perfect for proof-of-concept, integration testing and early AI or research experiments.
Built in Portugal on top of international quantum research and a resilient multi-source entropy architecture.
Multi-Source Quantum Entropy: From Physics to API
Multiple independent quantum hardware setups continuously measure vacuum-driven fluctuations and photon shot noise. Their outputs are combined, conditioned and tested to remove bias and detect anomalies. The result is a resilient, low-outage entropy layer you can call over HTTPS – with every sample linked to a verifiable hash for later checks.
Why randomness is the weakest link
Most modern systems treat randomness as an implementation detail: a function in a standard library, a setting in an HSM, a box you tick in a CSP. But when your randomness is predictable, everything built on top of it becomes brittle — keys, seeds, draws, simulations and even the behaviour of AI systems.
PureRand is built on a simple belief: the origin of your entropy deserves the same scrutiny as your cryptography, infrastructure and compliance. If you can point to physical processes, describe how they are monitored and show how outputs can be verified, you are operating on a different level of assurance — whether you are protecting keys, training models or running medical simulations.
Instead of hoping a generic PRNG is “good enough”, you can document exactly how randomness enters your system — and prove it to auditors, customers, researchers and yourself.
What goes wrong with weak entropy
- Keys that look unique but can be guessed or replayed.
- Games and lotteries that are “random” until someone finds the pattern.
- AI systems that quietly inherit bias from a single PRNG source.
- Simulations that silently distort risk, pricing or clinical outcomes.
What PureRand gives you instead
- Entropy with a documented, multi-source quantum origin.
- Metadata and verify hashes for every response.
- A clear story to tell security teams, regulators and ethics boards.
- An API that is easy to integrate and hard to cheat.
Multi-source quantum randomness, explained
Science made entropy visible to a global audience. PureRand takes the same intuition – “derive randomness from the real world” – and implements it using multiple independently operated quantum setups. Each one measures quantum fluctuations in light, and together they feed a resilient, multi-source entropy layer.
We measure phenomena such as photon shot noise and vacuum-driven fluctuations: the fundamental uncertainty in how individual photons arrive at detectors. Those noisy signals are digitised, whitened using proven extraction techniques and continuously tested. Only then do they become part of the entropy pool that feeds your API calls.
Because entropy is aggregated from multiple independent quantum sources across regions and technologies, PureRand avoids a single point of failure. If one source is degraded or offline, requests are seamlessly served from healthy sources while maintaining entropy quality.
From your perspective as a developer, all of this complexity collapses into a set of simple endpoints. Behind them is a pipeline designed for one purpose: produce randomness that you can rely on in adversarial, high-stakes and bias-sensitive environments.
1. Quantum signals
Multiple quantum hardware setups capture photon shot noise and related vacuum fluctuations – purely quantum sources of randomness, independent of software tricks or timing side channels.
2. Conditioning & tests
Raw measurements are digitised, whitened and subjected to continuous health tests. If something looks off, the signal from that source is quarantined instead of silently degrading your entropy.
3. API & verification
Each response includes metadata and a verify hash. Depending on your plan, entropy can also be anchored in a tamper-evident ledger for future audits and fairness checks.
4. Observability
The dashboard gives you insight into usage, health metrics, routing and latency. You can show, not just tell, how multi-source randomness is managed in your system.
Where PureRand fits in your architecture
PureRand is not a generic “random endpoint”. It is an entropy service designed to sit next to your existing cryptography, HSMs, agents, research pipelines and engines. Use it where the quality, origin and bias properties of randomness actually matter — and keep using your local PRNGs where they do not.
Cryptography & keys
Use PureRand when keys or secrets are generated — the moments where a bad seed compromises everything.
- Wallet seeds and private keys
- API keys and token seeds
- HMAC / signing secrets and challenges
Bias-aware AI & sampling
Inject physical, multi-source randomness into sampling loops and agent policies to reduce hidden bias.
- LLM sampling and diversity
- Decision-making for autonomous agents
- Stochastic search and exploration
Gaming & fairness
Make it provable that a draw, spin or shuffle was not influenced by anything except chance.
- Provably fair lotteries and draws
- On-chain randomness for smart contracts
- Player-visible verification tools
Scientific & medical research
Backtest strategies and run simulations with a defensible, verifiable entropy source.
- Monte Carlo risk and clinical models
- Epidemiological and medical simulations
- Scientific, pricing and stress testing
Try the API in under a minute
You do not need a new stack to use PureRand. A single HTTP request already gives you quantum-grade, multi-source randomness with metadata and a verify hash you can log or feed into your own checks.
The free tier gives you 5 calls per month. That is plenty to integrate PureRand, run a few end-to-end flows and decide whether you want to rely on it for real workloads.
For full endpoint details, parameters and examples, visit the API documentation.
# Request 32 bytes of quantum randomness (hex-encoded)
curl "https://api.purerand.com/random?bytes=32&format=hex" \
-H "Authorization: Bearer <YOUR_API_KEY>"
# Example response:
{
"source": "multi-source-quantum",
"bytes": 32,
"format": "hex",
"data": "7f90bbde0dce2a2a2ae19bc8cf8ad9f106f4b7a243e2dd7c1ad9f2c78fe2e91",
"verify": "/qr/2fb1c3d0e8d...8731",
"timestamp": "2025-12-18T14:03:12Z"
}// Using fetch in Node or modern browsers
const res = await fetch(
"https://api.purerand.com/random?bytes=16&format=base64",
{ headers: { "Authorization": "Bearer <YOUR_API_KEY>" } }
);
if (!res.ok) {
throw new Error("PureRand error: " + res.status);
}
const body = await res.json();
console.log("Random bytes:", body.data);
console.log("Verify hash :", body.verify);
// Log 'verify' alongside model seeds, parameters or experiment IDs
// to support reproducibility, fairness checks and future audits.Pricing that follows your risk, not your curiosity
Use the free tier to explore PureRand and connect it to your stack. When randomness becomes a real dependency — for key generation, fairness, bias-aware AI or risk modelling — we can scale with you.
Starter
Perfect when you are validating ideas, running small experiments or simply curious about multi-source quantum randomness as a service.
€0
per month
- 5 API calls / month
- 1 project / API key
- Access to dashboard
- Basic email support
Pro
For teams who want PureRand as a standard component in their security, gaming, AI or research pipelines, with usage and support aligned to production stakes.
Custom
usage-based
- Generous request volume
- Multiple projects & API keys
- Priority routing and monitoring
- Access to dashboard & analytics
Enterprise
For organisations that treat entropy as part of their regulated surface: dedicated capacity, custom integrations and support for deep audits and scientific or medical use-cases.
Custom
annual
- Dedicated quantum entropy channels
- On-prem / HSM integration options
- Extended audit & compliance support
- Priority SLA & support
About PureRand
PureRand is an independent technology project built in Portugal with a simple focus: make high-assurance, multi-source randomness accessible as a clear, well-documented service. We sit at the intersection of physics, security engineering and product design.
The quantum hardware that powers PureRand – lasers, vacuum setups and measurement systems – lives in specialised environments. Around it, we build software, observability and documentation so you do not need to become an experimental physicist to use a quantum entropy source.
Development of PureRand has been supported under the Portuguese PRR (Plano de Recuperação e Resiliência), specifically the “Vouchers para Startups – Novos Produtos Verdes e Digitais” call. Within this programme, PureRand contributes as a digital product that strengthens security, transparency, fairness and the responsible use of quantum technologies in real systems.
For more detail on scope, milestones and technical documentation related to the PRR voucher, see the PRR overview page or contact us directly.
How to talk to us
We work with security teams, founders, researchers and platform owners. Most conversations start with a short call where we map your use case to the parts of PureRand that actually matter for you.
- Security & architecture reviews
- Proof-of-concept integrations
- PRR / innovation programme alignment
- Research, medical and education collaborations
For general questions, partnerships or media, you can reach out:
use the contact form on /contact.