Versions
v2.13.12026-11-14patch release

Retry now uses the model you just picked

Here is a habit a lot of people have. You send a message, the reply starts coming in, and two lines in you can tell it is the wrong model for the job. Maybe you wanted something faster, maybe you wanted a model that is better at code, or maybe you just made a typo and want a clean answer. So you press Stop, open the model picker, choose a different model, and press Retry on the reply you stopped. Until now that Retry quietly went back to the model you had before the switch, so you got the same kind of answer again and had to send your message a second time to actually reach the new model. That was a bug, and it is fixed. Retry now always asks the model that is selected in the picker at the moment you press it. The same goes for editing an earlier message: when you change the model and then edit a message to resend it, the new reply comes from the model you chose, not the old one. Your effort setting and Auto mode follow the same rule, so whatever the composer shows when you press Retry or save an edit is exactly what the request uses. The reply you stopped is replaced as before, and everything earlier in the conversation stays where it was. Nothing changes if you never switch models, and you do not need to refresh the page or start a new chat to get this fix. It works the same on the web, in your phone's browser and in every language Portail supports, including Russian.

640 words

Stop a reply, switch models, press Retry: the new model answers.

Editing a message after switching models also uses the new model.

Effort and Auto mode are read fresh on every Retry and edit.

Portail v2.13.1 shipped on 2026-11-14 as a patch release focused on retry now uses the model you just picked. Here is a habit a lot of people have. You send a message, the reply starts coming in, and two lines in you can tell it is the wrong model for the job. Maybe you wanted something faster, maybe you wanted a model that is better at code, or maybe you just made a typo and want a clean answer. So you press Stop, open the model picker, choose a different model, and press Retry on the reply you stopped. Until now that Retry quietly went back to the model you had before the switch, so you got the same kind of answer again and had to send your message a second time to actually reach the new model. That was a bug, and it is fixed. Retry now always asks the model that is selected in the picker at the moment you press it. The same goes for editing an earlier message: when you change the model and then edit a message to resend it, the new reply comes from the model you chose, not the old one. Your effort setting and Auto mode follow the same rule, so whatever the composer shows when you press Retry or save an edit is exactly what the request uses. The reply you stopped is replaced as before, and everything earlier in the conversation stays where it was. Nothing changes if you never switch models, and you do not need to refresh the page or start a new chat to get this fix. It works the same on the web, in your phone's browser and in every language Portail supports, including Russian. The work is recorded here because small releases are still product history. They change how people sign in, read responses, pay for access, share conversations, open files, or trust the system when something goes wrong.

The main changes were direct: stop a reply, switch models, press retry: the new model answers, editing a message after switching models also uses the new model, and effort and auto mode are read fresh on every retry and edit. That list is intentionally plain. It names what changed without turning the release into theatre. When a release fixes a build, a redirect, a storage path, or a UI edge case, the value is often that users stop noticing the problem. The archive keeps that invisible work visible enough for future debugging and for anyone checking whether Portail has cared about the details over time.

This version also fits into the wider Portail contract. One account, one chat history, and one model picker only work if the smaller pieces keep their promises. Authentication has to fail clearly. Billing has to count honestly. Uploads have to arrive in a form models can read. Admin tools have to show the operator what happened instead of hiding it behind a blank screen. v2.13.1 moved that contract forward in its own narrow lane.

The screenshots attached to this entry show the public release record as it now appears on portail.cc. For older releases, this archive was rebuilt from the git history, package version changes, release modal text, and commit subjects. That means the page is honest about its source: it documents the shipped change and preserves the record going forward, even when the exact old production screen cannot be recreated from a single saved image.

The practical result is simple. Someone landing on this page can see the version number, date, release tier, headline, highlights, and a real rendered screenshot of the archive page. Search engines can index the post through the canonical URL, sitemap entry, and structured data. Future releases now have a standard to meet before deploy, so the history should get cleaner from here instead of fuzzier.