# Archetype x zkSecurity - Proof is in the Pudding: zkML

- **Authors**: ZK/SEC
- **Date**: September 30, 2026
- **Tags**: educative, zk, zkml
https://www.youtube.com/watch?v=yaqDe1yXvjA

For the 11th session of Proof is in the Pudding, we teamed up with [Archetype](https://www.archetype.fund/) to talk about **zkML**. Running an LLM is already expensive. How much harder is it to prove that you ran it?

We start with what a proof of model inference actually guarantees, how inference differs from training, and the alternatives to using ZK. Then we open up a transformer and look at the computation we would have to prove.

From there, we discuss sumcheck, GKR, lookup arguments, and provers specialized for particular models. If you'd like to work through one of those building blocks yourself, our [sumcheck tutorial](https://blog.zksecurity.xyz/posts/sumcheck-tutorial/) includes code and exercises. We also covered GKR from a security angle in [Session 3](https://blog.zksecurity.xyz/posts/pudding3/).

The second half looks at quantization, models designed to be easier to prove, and opportunities around KV caching. We finish with scaling, where verifiable ML might be useful, and a more speculative possibility: agents agreeing on circuits and exchanging proofs as they interact.

## Jump to a topic

- [1:44: What is zkML?](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=104s)
- [5:32: Inference vs. training](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=332s)
- [7:52: Alternatives to ZK for verifiable inference](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=472s)
- [10:44: Inside a transformer](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=644s)
- [15:25: Proving transformer computation](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=925s)
- [27:12: Quantization and SNARK-friendly models](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=1632s)
- [31:58: KV caching and optimization opportunities](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=1918s)
- [35:08: Agent-to-agent proofs and dynamic circuits](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=2108s)
- [37:18: How zkML scales](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=2238s)
- [38:50: The market for verifiable ML](https://www.youtube.com/watch?v=yaqDe1yXvjA&t=2330s)

You can catch up on [our previous session on Groth16](https://blog.zksecurity.xyz/posts/pudding-10-groth16/) or browse [the full series](https://blog.zksecurity.xyz/series/proof-is-in-the-pudding/). Have a topic you'd like us to cover next? Let us know on [Twitter/X](https://x.com/zksecurityXYZ)!

---

This article was published on the [ZK/SEC Quarterly](https://blog.zksecurity.xyz) blog by [ZK Security](https://www.zksecurity.xyz), a leading security firm specialized in zero-knowledge proofs, MPC, FHE, and advanced cryptography. ZK Security has audited some of the most critical ZK systems in production, discovered vulnerabilities in major protocols including Aleo, Solana, and Halo2, and built open-source tools like [Clean](https://github.com/Verified-zkEVM/clean) for formally verified ZK circuits. For more articles, see the [full list of posts](https://blog.zksecurity.xyz/llms.txt).