# Pearl research library

https://www.learnpearl.org/research/ · Checked October 20, 2026 · Advanced

Pearl makes claims across cryptography, distributed systems, AI infrastructure and economics. No single document settles all four. The fastest way to a serious view is to read the protocol claim, the running code and the strongest critique together.

## Key takeaways

- Two whitepapers exist: the INT paper describes the live network; “The Pearl Protocol” specifies an FP8 version that isn’t live.
- The cryptography, the integration, the observed usefulness and the economics are four separate claims.
- A sound proof can coexist with weak adoption, and vice versa.

## Recommended reading order

1. **The INT whitepaper.** The scheme mainnet runs: protocol design, the proof-of-useful-work construction, parameters, addresses and issuance. [Read the INT whitepaper](https://pearlresearch.ai/research/int-whitepaper).
2. **Proofs of useful work from arbitrary matrix multiplication.** The 2025 paper by Ilan Komargodski and Omri Weinstein that the repository names as Pearl’s basis: the math that lets chosen matrix multiplication serve as proof of work. [arXiv 2504.09971](https://arxiv.org/abs/2504.09971).
3. **The open-source implementation.** What actually ships: pearld, Oyster, the light client, the proof system and the vLLM miner. [GitHub](https://github.com/pearl-research-labs/pearl).
4. **The usefulness-gap study.** An empirical challenge to whether observed mining was useful AI work, measured in late May 2026. [arXiv 2606.04819](https://arxiv.org/abs/2606.04819).
5. **Pearl’s explainer and benchmarks.** Pearl’s own walk-through of its approach, with its measured serving overhead. [Proof of Useful Work from the Ground Up](https://pearlresearch.ai/research/blog/proof-of-useful-work).
6. **The economics.** When block rewards can increase the supply of useful compute without making attacks cheap. [arXiv 2606.06700](https://arxiv.org/abs/2606.06700).
7. **An outside systematization.** An independent reading of Pearl’s proof-of-useful-work scheme that proposes letting each miner choose its own noise level. [arXiv 2610.25347](https://arxiv.org/abs/2610.25347), October 2026.
8. **What comes next: FP8.** “The Pearl Protocol” specifies an FP8 floating-point version (not live), drawing on [Hawkeye](https://arxiv.org/abs/2603.20421) for reproducing GPU arithmetic exactly. [Read the spec](https://pearlresearch.ai/research/whitepaper).

## Four claims to keep apart

- **Cryptographic:** Can MatMul prove work? · Verification and security
- **Technical:** Can it share an AI workload? · Integration and overhead
- **Empirical:** Is current mining useful? · Observed real-world behavior
- **Economic:** Will useful compute create value? · Adoption, incentives and demand

> **A proof can be sound while adoption is early.** Weakness in today’s deployment doesn’t disprove the math. Likewise, proving the construction doesn’t guarantee commercial demand or decentralization.

## Questions worth tracking

- What fraction of network work is paired with real training or inference?
- How concentrated are miners, pools, proof generation and AI customers?
- What proof overhead and verification cost show up across current model architectures?
- How does the move to FP8 (certificate V4) change accuracy, hardware requirements and security?
- How does the network behave under fast swings in work and adversarial timestamps?
- Can wallets and exchanges mature without becoming central points of failure?

## Our source policy

Protocol facts come from the code and the official whitepapers, and where the two disagree, the code wins. Release facts come from the project’s GitHub releases. Performance and adoption claims are labeled by source and date. Third-party tutorials are never treated as protocol specifications.

When credible sources disagree, we say so and name the evidence that would settle it. A project claim doesn’t become a fact by being repeated.

## Sources

- [Pearl INT whitepaper (the scheme mainnet runs)](https://pearlresearch.ai/research/int-whitepaper) (Pearl Research Labs, primary)
- [“The Pearl Protocol”: the FP8 specification (not live)](https://pearlresearch.ai/research/whitepaper) (Pearl Research Labs, primary)
- [Proofs of Useful Work from Arbitrary Matrix Multiplication (arXiv 2504.09971)](https://arxiv.org/abs/2504.09971) (arXiv, research)
- [The Usefulness Gap in Proof-of-Useful-Work: An Empirical Study of Pearl’s cuPOW Protocol (arXiv 2606.04819)](https://arxiv.org/abs/2606.04819) (arXiv, independent)
- [The Economics of Proof-of-Useful-Work (arXiv 2606.06700)](https://arxiv.org/abs/2606.06700) (arXiv, research)
- [Hawkeye: Reproducing GPU-Level Non-Determinism (arXiv 2603.20421), by researchers at Pearl Research Labs and Stanford](https://arxiv.org/abs/2603.20421) (arXiv, research)
- [On Pearl Blockchain’s Proof of Useful Work: Systematization and Flexibility (arXiv 2610.25347)](https://arxiv.org/abs/2610.25347) (arXiv, research)
- [Proof of Useful Work from the Ground Up](https://pearlresearch.ai/research/blog/proof-of-useful-work) (Pearl Research Labs, primary)
- [Pearl monorepo on GitHub](https://github.com/pearl-research-labs/pearl) (Pearl Research Labs, primary)