Anthropic Open-Sources FlashProve: Verifying Model Outputs with Zero-Knowledge Proofs
English text is being prepared.
Anthropic has announced its verifiable inference research FlashProve, enabling developers to verify whether a given response came from a specified model, and has open-sourced the code. The solution is based on a zkSNARK architecture: model outputs come with a mathematical proof attached, and users can complete verification on an ordinary CPU within seconds without leaking model parameters. According to reports, on a 400B MoE Llama 4, a single H100 can generate proofs for 110 tokens per second, with proof files of about 19MB; the code and the Lean formal security proof are open-sourced under the Apache 2.0 license.
Why read this
AI score 74
Anthropic has open-sourced a verifiable inference solution, allowing readers to learn how to use cryptographic proofs to verify whether model outputs come from specified weights.
Topics
View original postNew Anthropic research: progress on provable inference for large language models. Here, we introduce FlashProve. Building on established cryptographic techniques, FlashProve allows developers to verify whether a given response came from a given AI model. It’s substantially more efficient than prior open-source methods for doing so. These proofs have a range of applications. They can, for instance, make it harder for attackers to hijack a chat, or enable safety evaluators to verify that the model they’re testing is the same one being served to customers. For our full write-up, see here: https://t.co/OIZWPb6QZa Further progress in provable inference will likely be an important part of improving trust in AI deployment. To that end, we're open-sourcing FlashProve’s code, here: https://t.co/7ky3Ir8V06 We’re grateful to our collaborators: Alessandro Chiesa, @GiacomoFenzi, @gwberg95, @miranda__christ, Rahul Dalal (https://t.co/pjWKLHWXyz) and Ben Mayer (@transmissions11).
View original
New Anthropic research: progress on provable inference for large language models. Here, we introduce FlashProve. Building on established cryptographic techniques, FlashProve allows developers to verify whether a given response came from a given AI model. It’s substantially more efficient than prior open-source methods for doing so. These proofs have a range of applications. They can, for instance, make it harder for attackers to hijack a chat, or enable safety evaluators to verify that the model they’re testing is the same one being served to customers. For our full write-up, see here: https://t.co/OIZWPb6QZa Further progress in provable inference will likely be an important part of improving trust in AI deployment. To that end, we're open-sourcing FlashProve’s code, here: https://t.co/7ky3Ir8V06 We’re grateful to our collaborators: Alessandro Chiesa, @GiacomoFenzi, @gwberg95, @miranda__christ, Rahul Dalal (https://t.co/pjWKLHWXyz) and Ben Mayer (@transmissions11).
Body
那些无良AI套壳中转站完蛋了， Anthropic 官方直接给出了照妖镜： Claude大模型防伪证书来了！ 以后再也不用担心AI中转站暗中用低价模型偷偷套壳赚取暴利了，
Anthropic 正式开源 FlashProve：400B 巨模实现每秒 110 token 极速零知识证明，彻底消灭假冒接口，
大家平时调用各种大模型 API 或者第三方中转站时， 最头疼的一件事就是根本无法验明正身： 中转商收着顶配模型的钱，背地里偷偷换成廉价的小模型套利，你看不出来； 黑客劫持了数据流篡改答案，你从文本表面也看不出来。
今天 Anthropic 丢出来的这套系统，直接给每一次模型调用盖上了不可伪造的密码学钢印。
它基于零知识证明（zkSNARK）架构， 大模型在输出回答的同时，会附带一张极小的数学证明， 你在自己的普通电脑 CPU 上几秒钟就能完成验真， 既能 100% 确认这行字确实来自那套特定的官方权重，又绝不泄露大厂任何底层的商业机密参数。
最炸裂的是它的工程性能： 在 4000 亿参数（400B MoE）的 Llama 4 巨模上，单张 H100 显卡每秒能生成 110 个 token 的证明，证明文件仅约 19MB； 相比两年前开源方案动辄十几分钟跑一个 token 的龟速，速度直接飙升了几个数量级， 整套代码和 Lean 形式化安全证明直接走 Apache 2.0 协议全量开源。
以前调用接口全凭中转商的良心， 从今天起，每一次 Token 生成都有了不可抵赖的数学收据。
谁在偷换模型套利，谁在拿小模型假冒旗舰，只要把这套验证器挂上去，所有的猫腻当场无所遁形。
开源仓库与核心论文整理在一楼了。