


Your provably private way to access all of the latest and greatest AI models and tools. Bitcoin, Crypto, and Cards. Pay-per-use, no subscriptions!


Revenue on our @TinfoilAI-hosted TEE models is up ~20x since launch in March. TEE models make it so that nobody, not us, not the model host, can read your prompts. Private inference won't be a niche. It's going to become the default.

On the surface this sounds great. I am all for paying with bitcoin. But lightning micropayments at scale are slow and not 100% reliable, leading to a worse user experience than you get today from standard inference. There is already some trust involved between the buyer and seller, I think @PPQDotAI with automatic @nwc_dev bitcoin topups is a better UX. Micropayments are cool in theory but what problem do they solve here? can you think of any new usecase?
Today we're announcing pi402! Pay for AI inference directly with your coding agent. Instant on Lightning. No accounts, API keys, or cards needed. Inference 🤝 Bitcoin Learn more: pi402.tech

We @PPQdotAI use a TEE router. It makes it so that we: 1) Cannot see the content of our users’ queries 2) Prove the model and provider that the query ended up executing (no model fraud possible) Other routers follow the “trust me bro” architecture.
Trusted Execution Environments Are Cypherpunk Magic with Hillebrand & Ben Perrin.



Voxtral TTS is now available on PPQ! High quality, cheap TTS served inside of @TinfoilAI's secure enclaves. Privacy assured!



Updating to 2.4.7 on Thursday saved us. Stay on top of updates folks!
📢 Security Incident Affecting BTCPay Server 2.4.1 and Earlier Read the full security advisory, including details & timelines, on our blog: blog.btcpayserver.org/btcpay… If you were impacted, please send all details and logs to security@btcpayserver.org


New guide: Strix + PayPerQ. A project with 70k Github starts that pentests your app. Swap in one of our new abliterated models when proprietary models refuse cyber questions. ppq.ai/blog/using-strix-with…

Been working on post training models and got some cool results. Was able to train Qwen3.8:27b to be significantly better at working with bitcoin scripts.

Many customers asking how to red-team their own codebase. We just added the @abliteration_ai models (GLM 5.3 and DeepSeek V4.1 Flash, abliterated) Refusals are removed from the weights, so they work through your security issues instead of declining! ppq.ai/blog/introducing-abli…

Most people overpay for AI 🤯 @PPQdotAI: hundreds of models, uncensored ones included. No account, no subscription. Pay per prompt with Lightning. It even runs my Hermes agent, with auto top-ups from @getAlby. Full tutorial on @SOVsessions 👇 youtu.be/Er1MIpJ8uHw
