✨What's New

Until now, you couldn’t see which model powered your agent, or choose another. Now Claude Sonnet 5.5 is the default, and in an agent’s settings you can pick Claude Opus 5.5, the most capable, or try an experimental model. You can see which model your agent runs on, and new jobs name the model they used. Once an agent is at work, its settings are simpler too.

• Claude Sonnet 5.5 is the new default

  Every agent now runs on Claude Sonnet 5.5 by default, up from Claude Sonnet 5, in chats and jobs alike. Your existing agents move over on their own, with nothing for you to change.

• Choose the model for each agent

  In an agent’s settings, pick the model it runs on: Claude Sonnet 5.5 is balanced and recommended, and Claude Opus 5.5 is the most capable. Each model shows a simple $ to $$$ cost guide.

• Try other models, clearly marked Experimental

  Models other than Claude sit apart under Experimental models and carry an Experimental tag wherever you pick them, so you know when you’re trying something less proven.

• See which model is doing the work

  An agent at work shows its model in the top bar; tap it for details and a shortcut to change it. Once a new job finishes, your Jobs list names the model it ran on.

• Simpler settings, and a setup that stays put

  Once an agent is at work, its settings come down to Model and Routines. The setup its maker built stays put, and the chat can no longer rewrite it.

Once a job had started, you couldn’t stop it, and if it broke partway through, its card could say “Working” for hours. Now a Stop button ends a running job, jobs start on a fresh machine of their own, and your agent takes one job at a time. Cards say what really happened, and the chat opens at your newest message.

• Stop a running job with one tap

  Tap Stop on a running job’s card and the work ends, shown calmly as Stopped, not as a failure. If Slate can’t confirm the job has stopped, it says so, and you can try again.

• A fresh machine, one job at a time

  Jobs used to share a machine, so one could cut another short. Now jobs start on a fresh machine of their own, and your agent takes one at a time, saying when it’s busy.

• An honest result when a job breaks

  If a job stops unexpectedly partway through, its card says so within 10 minutes, not hours later, and the files it saved are kept. A job that fails to start now shows as failed.

• Jobs without steps get a name and status

  A quick job without steps used to sit empty in your Jobs list and never reach the Board. Now it shows one line with its name and status: working, done, failed, or stopped.

• Chat opens at your newest message

  Opening a chat could leave it blank or stuck partway up the conversation. Now it opens at your newest message and holds there while job cards and pictures load, until you scroll up yourself.

Small things could trip you up between signing in and getting your work back: an image that never reached the job, a video that wouldn’t play, a file silently ignored, or a vague error when credits ran out. Each of these now works, or tells you plainly why it can’t. Signing in is safer too, and only you can remove messages from your chats.

• Your chats and sign-in are better protected

  Only you can remove messages from your chats. Sign-in and sign-up forms now wait until the page is ready before you can type, so your email and password never appear in the address bar.

• Images you add to chat reach the job

  Drop or paste an image into the message box, and a job started from that request now works from your actual picture. Before, it could show in the chat but never reach the job.

• A clear reason when a file won’t attach

  Drop or paste a file the message box can’t take, like one that isn’t an image or is too large, and a note now says which file and why. Your typed message stays put.

• Delivered videos open and play

  A delivered video could show “Video unavailable” even though the video itself was fine. Now it opens and plays.

• A plain answer when your credits run out

  If a reply stops because you’ve run out of credits, the chat now says so and suggests upgrading your plan, instead of a bare “An error occurred.” You know what to do next.

A long job used to depend on your open page, and it stopped after about half an hour. Now jobs run on their own for up to six hours, keep going if you close the tab, and save their result to the chat when they finish. Come back partway through and the job card catches up.

• Jobs keep going for hours

  A job can now run for up to six hours, so bigger pieces of work no longer stop at the half-hour mark.

• Close the tab, and the work carries on

  Closing the tab, losing your connection, or an update to Slate no longer cuts a job short. It keeps working and saves its result to the chat when it’s done.

• Come back to a job in progress

  Reload the page or return later, and the job card catches up with what the job is doing, then keeps updating as the work continues.

Slate used to work well only on a computer. On a phone, the chat shrank to a sliver and some buttons appeared only under a mouse. Now every page fits phones and tablets. Setting up an agent shows each step as it happens, a failed image or video no longer costs credits, and a failed payment is explained plainly.

• Every page fits your phone and tablet

  A menu button opens the sidebar on every page, and on an agent’s page you switch between its work and the chat. Buttons that needed a mouse now show on touch screens.

• Files open full screen on small screens

  Open a file on your phone and it fills the screen, with a large close button. Typing a message no longer makes your phone zoom in.

• Watch your agent being set up

  Setting up an agent used to mean a small spinner and a silent wait. Now each step shows as it happens, including each skill being added, and you can try again if it fails.

• Failed images and videos are free

  If an image or video your agent asks for fails to generate, you’re no longer charged for it. Your agent gets the real reason, not a false success.

• A clear message when a payment fails

  If a renewal payment fails, Slate now says so plainly, with an Update payment button that takes you straight to fixing it. Once it’s paid, your access returns on its own.

Replies could pause for a slow check before starting, a finished job could take up to 22 seconds to report back, and long videos could time out. Now every agent runs on Claude Sonnet 4.6, replies and results come sooner, and long videos run to the end. When something goes wrong, the error stays on screen with a way to fix it.

• Every agent runs on Claude Sonnet 4.6

  Chats and jobs alike now run on Claude Sonnet 4.6, up from earlier Sonnet models, with nothing for you to change.

• Faster replies, faster results

  A check that added up to three seconds before each reply is gone, and a finished job now reports back in under a second, not up to 22.

• Errors stay on screen, with a fix

  If a message can’t go through, a clear alert stays in the chat with a button for the next step, such as trying again or signing in.

• Long videos run to the end

  Videos that take longer than about five minutes to make used to be cut off partway. Now they keep going until the video is ready.

• FixTasks stuck after a job ends

  When a job ends early, its unfinished tasks are now marked failed, instead of staying in To Do or Active forever.