The best AI solution Bitcoin & Nostr has to offer. We want to build a solution so good it'll bring a new influx of users to Nostr and Bitcoin. View quoted note →

New Release! Routstr Core v0.4.4: Reliable Billing & Robust Streaming Major Changes: - **More Accurate Cache Pricing:** Cached input tokens are now billed at their real, per-vendor rates across direct upstreams (OpenAI, Anthropic, DeepSeek, and more) instead of a flat guess. Cache prices are configurable, so operators can tune them per model. - **Reliable Mint Swaps:** Foreign-mint swaps now retry using the fees the mint actually charges instead of trusting a stale `fee_reserve`, and the NUT-02 input fee is deducted when crediting trusted-mint top-ups. - **Robust Streaming for All Providers:** A new buffer-based SSE parser handles streamed responses for every supported provider - **Lightning Invoice Top-ups (RIP-08):** You can now create and top up API keys directly with a Lightning invoice. - **Hardened Child Keys:** Child key balance deduction. Plus better rate-limit forwarding, smarter OpenRouter defaults (routes to a tool-use-capable provider), harder-to-abuse admin endpoints, and many smaller fixes. Huge thanks to everyone who shipped this release - with a special welcome to our first-time contributors. Please upgrade your nodes by pulling from the release tag here:

GitHub

Release Routstr Core v0.4.4 · Routstr/routstr-core

What's Changed configure payment by @9qeklajc in #494 fix provder forwarded path & wrong html response by @9qeklajc in #496 fix-docker-build by @9...

and run docker compose up --build -d. If you want to offer Tinfoil models as well, you can checkout the integration tag here (still in Beta):

GitHub

Release v0.4.6-e2ee · Routstr/routstr-core

Release v0.4.6-e2ee

If you're running this node, please reach out. You've misconfigured some things in your node. We will make sure to prevent this from happening in the future through better node runner UX 😅 View quoted note →

Big Release for Private AI 🔒 TEE E2EE models are live on Routstr! Access frontier open-source models like GLM-5.2 and DeepSeek V4 Pro inside verified GPU TEEs, encrypted on your device and only decrypted inside a TEE while you pay per prompt in sats with no accounts. The best privacy you'll get apart from running them yourself! Write code anonymously with Opus-like capabilities via routstrd, or ask private questions on our chat app. Read more about how it works: View article →

GLM-5.2 is the best open-source model available right now. And it **beats** GPT-5.5 in many benchmarks. It is available on the Routstr network offered by 6 Routstr nodes already:

Browse AI Models | Routstr

Compare prices and features of top AI models including Llama 3, Mixtral, and more. Pay with Bitcoin Lightning.

It should go without saying that all new models popup immediately on Routstr. Happy hacking on your Nostr projects!

Yep. You can run Routstr daemon for your team or for multiple devices of your own by hosting it on Cloudron or just using docker on a VPS. Setup NWC to automatically top up using Lightning and access more than 300+ models including all SOTA models soon as they're launched. Bitcoin, Nostr and Cashu/Lightning View quoted note →

@Thefux from Routstr is in Prague demoing the Routstr ecosystem. Be sure to check it out and set yourself up with the sovereign AI stack. View quoted note →

Routstr network just surpassed 2.1M sats in monthly volume for the first time! Next goal: 21M sats in monthly volume. 💹 *Fully organic, grass fed, no hype. Slow but steady.