# Lean proof assistant

When a user queries a large model, nothing in the reply shows it came from the model the developer claims, so a routing service swapping in a cheaper model or an attacker tampering with the token stream goes unnoticed. Anthropic's open-source FlashProve attaches a zero-knowledge succinct argument (zkSNARK, a short certificate a verifier can check quickly without the weights) to inference. On 4-bit Llama 4 Maverick, a 400B-parameter mixture of experts, a single H100 generates proofs at about 110 tokens per second, roughly 19 MB per 1,024-token request, at about 70 to 95 times the GPU time vLLM needs to run the model itself.

When a user queries a large model, nothing in the reply shows it came from the model the developer claims, so a routing service swapping in a cheaper model or an attacker tampering with the token stream goes unnoticed. Anthropic's open-source FlashProve attaches a zero-knowledge succinct argument (zkSNARK, a short certificate a verifier can check quickly without the weights) to inference. On 4-bit Llama 4 Maverick, a 400B-parameter mixture of experts, a single H100 generates proofs at about 110 tokens per second, roughly 19 MB per 1,024-token request, at about 70 to 95 times the GPU time vLLM needs to run the model itself.

The classification of finite simple groups — the longest mathematical proof humans have written — now exists as a complete, computer-checked proof. That "enormous theorem" had been spread across more than 500 articles and roughly 15,000 pages, written over decades by about 100 mathematicians. Engineers and mathematicians at Axiom Math directed swarms of dozens of AI agents for 24 days, producing about 43.1 million lines of Lean code, and recorded 34 noteworthy deviations from the published record, the largest of which concerned the Unbalanced Group Theorem; the process also yielded a new theorem in group theory.

The classification of finite simple groups — the longest mathematical proof humans have written — now exists as a complete, computer-checked proof. That "enormous theorem" had been spread across more than 500 articles and roughly 15,000 pages, written over decades by about 100 mathematicians. Engineers and mathematicians at Axiom Math directed swarms of dozens of AI agents for 24 days, producing about 43.1 million lines of Lean code, and recorded 34 noteworthy deviations from the published record, the largest of which concerned the Unbalanced Group Theorem; the process also yielded a new theorem in group theory.