The Anyroute Network is live. $ANYR Private inference should not depend on a handful of opaque providers. More private capacity needs to be operated by more independent hosts, with every server proving what it is running. The Network opens that supply layer. Eligible operators can run a SEAL host on confidential GPU or TDX infrastructure, connect it to Anyroute, and serve private inference through the network. What is included: • One-command host setup and a readiness check • Hardware attestation before a host can serve private traffic • Model-weight and server configuration verification • Attested routing, health checks, and automatic failover • Signed receipts for every response • Public, attested, and unlinkable privacy lanes • Per-lane status, SLOs, and incident feeds • Token-based USDG settlement for capacity served • Host bonding, dispute windows, and slashing rules for bad actors • A public network page to check whether your infrastructure qualifies For users, it means more verified private capacity and less dependence on a single endpoint. For operators, it means private hardware can become part of a verifiable inference network. The goal is not just more GPUs. It is private inference with a record: what model ran, where it ran, what was verified, and what happened on every call. Check your server: anyroute.tech/network Build: github.com/AnyRouteRH/AnyRou…

Website update: Railway has restored deployments, but we are keeping Anyroute offline while we move to a private VPS, rotate credentials, and test the setup. Target is 5–7 days. We’ll update you. Thanks for sticking with us.

Been a little quiet on here. No big speech. We are back at it, and we will have more to show soon.

Set the line before you leave. Choose the balance you want to keep. When your account drops below it, your agents pause themselves instead of burning through what is left. anyroute.tech github.com/AnyRouteRH/AnyRou…

If an agent is doing the same approved task for the next hour, you should not have to approve every single request. Approve it once for the hour. Your other limits still apply, and you can end the approval early whenever you want. anyroute.tech github.com/AnyRouteRH/AnyRou…

The fastest way through the product is often the keyboard. Press `?` on any page to see the shortcuts available there, with the right keys for Mac or PC. anyroute.tech github.com/AnyRouteRH/AnyRou…

Your agent can be stopped from your phone too. In your linked Telegram chat, type `/stop Shop bot 1h` to pause it for an hour. Use `/start Shop bot` when you are ready to start it again. anyroute.tech/docs/#telegram…

The full model catalogue is useful. Your own usage is more useful. Filter it to the models you have actually called, then see your spend and success rate for each one. anyroute.tech github.com/AnyRouteRH/AnyRou…

A long chat can quietly add up. Set a spending cap on a chat. A send that would go over it is held, so you can raise the cap or send anyway. anyroute.tech/docs/#chat-cap

A useful answer should not disappear into a long chat. Pin any message to the top, then jump back to it whenever you need it. anyroute.tech github.com/AnyRouteRH/AnyRou…

Some of the quieter work has been landing too: 1. Pin important messages in long chats 2. See the models you actually use, with spend and success rate 3. Press ? to see shortcuts for the page you are on More updates will be up soon. anyroute.tech

Your agent does not need access around the clock. Set its working days, hours, and time zone in Agents. Outside that window, it either blocks the request or sends it to you for approval. anyroute.tech/docs/#work-hou…

A request gets blocked. Instead of guessing, Anyroute tells you why. Every refusal links to the rule behind it and the exact setting you need to change. anyroute.tech github.com/AnyRouteRH/AnyRou…

When something is not working, it helps to know whether it is your setup or ours. Turn on incident alerts in Settings and get a notice when an issue opens, is updated, and is resolved, in your Inbox or Telegram. They are off by default. anyroute.tech/docs/#incident…

Before changing an agent’s rules, you should know what that change will actually do. Anyroute shows the difference in plain words and replays it against last week’s calls before you save anything. anyroute.tech github.com/AnyRouteRH/AnyRou…

You may be paying more than you need to for work a similar model can handle. Ways to save compares your last 30 days of usage with similar models and shows the estimated difference. Try one in Chat before you change anything. anyroute.tech/docs/#savings-…

A quick view of your account, without opening the dashboard. Choose a daily or weekly summary of your agents, spend, and models, then pick when it arrives. anyroute.tech github.com/AnyRouteRH/AnyRou…

When a model you rely on goes down, refreshing its page is not a plan. Watch it once and Anyroute lets you know when it becomes unavailable, then when it is back, in your Inbox or on Telegram. anyroute.tech/docs/#model-wa…

Some things need your attention, just not right now. Snooze an Inbox notice for an hour, until tomorrow, or until Monday. It comes back unread at the time you chose. anyroute.tech github.com/AnyRouteRH/AnyRou…

An approval or denial is more useful when the agent knows why. Add a short note when you approve or deny a request, from the web or Telegram. Your decision and the note are sent back to the agent together. anyroute.tech github.com/AnyRouteRH/AnyRou…

Replacing a key should not take your app offline. Rotate it in Settings, then choose how long the old key stays active while you update your apps: one hour, 24 hours, or seven days. After that, it switches off automatically. anyroute.tech/docs/#rotate-k…