Bit of a small one today, but we wanted to fix the latency issues introduced last week!

For this Halloween, the spirits have haunted us and made us do something we've never done before!

You heard right: Everyone still in a trial, or who had a taste before November 1st, got a reset - and we extended them by a week. Which means everyone has until November 21st to try out Novelcrafter (again).

Some are hobbyists who only write on weekends, others have family obligations; we know everybody's time is precious. With November being arguably the most popular month for writing, we decided to give everyone the time they need to explore Novelcrafter.

Big thanks to all the people who have given their feedback and filled out the trial survey so far (and continue to do so)!

If you want to be visited by the spirits as well, why not try out this month's game of Tarot cards to inspire your next story?

Shuffle the deck and draw cards for the Hero(ine)'s Journey, your character arc, or for world building inspiration. The writing prompt is hidden in the details. Will you find a tale of love and loss or perhaps a twist of fate?

We've updated the word counting system when you're using an up-to-date browser. This means that we now support a much wider array of languages, since the actual counting is provided by your system itself. If you're on an older browser (e.g. Firefox Version older than 125), please update accordingly to make use of the new version.

While we tested this with our existing changes we did on September 5, please report any errors you might find with your language!

On certain images (e.g. ones with glitter, line art, or harsh angles) the image uploads (especially covers) were not as pretty as they can be - we updated the internals to get rid of any (pixelated) artifacts, but any previously uploaded images have to be re-uploaded if you want to take advantage of the new quality.

Today's update is a patch release for things that cropped up with all the editor updates:

Have you ever got a nice response in Chat, then saved it as a snippet and discovered all the pretty formatting has been lost? Yeah, we, too - too many times.

With today's update we added proper support for lists (numbered and bullets), headings (H1 to H3) and other formatting options to (almost) all text editors:

For people who want to spend a monthly price, instead of credits, we now added native support for Featherless AI and Infermatic. They mainly host NSFW-friendly models, so these are a good alternative if you're writing romance stories.

We also added a setting to change chat message flow for threads: You can either have messages start at the top, or bottom, which should help with when you want to read the answer as soon as it comes in, without having to wait until it stops scrolling.

Furthermore, the extraction feature now supports tags/labels as well, if you add them via square brackets to the name, like the following format: .

This release updates the system prompt model lists with the following additions and changes:

This update includes a couple fixes for when data can't be saved and reduce data loss on unreliable connections as well as improving detection around logged out users.

You can now export your chats just like you can export your codex or snippets! In addition, the novel export screen now has a section to include other aspects of your novel to get a single "full project" export of your novel.

This update makes word counting to be more in line with Scrivener. Both, Word and Google Docs (which we have used as a guideline so far) have issues counting words with em-dashes, punctuation and other special characters. The updated word counting is now more like what people would count as real words.

For comparison:

We internally upgraded to a new version of the framework Novelcrafter is using to render stuff on screen. While there were great performance improvements, the memory consumption should be even less for bigger novels now. Though we did extensive testing, as always, please report if some things aren't updating quite right.

For all system prompts, we now configured all Mistral models to go through the "Mistral" provider on OpenRouter. While OpenRouter does allow calls to "Azure" as well, those have a special moderation layer that can cut off the output too early if it detects an issue, causing you to loose money as the AI already started responding. Novelcrafter cannot show you when these "in-stream moderation" errors happen yet, so we opted for this solution for the time being.

This release mostly contains bug fixes and general improvements: