new

improved

fixed

Prompts

AI

October 1, 2025

In case you missed it, we also posted a community update blog post recently: https://www.novelcrafter.com/blog/2025-09-community-update
---
We haven't posted an update in a while, so here's what's new today (plus some fixes we've made over the past few weeks):
🤖 AI Connections
We've updated and enhanced the inner workings of all AI connections:
  • Added a new generic "OpenAI Compatible" provider to support other niche platforms. If we don't have official support for a vendor, you should now be able to use them through this connection!
  • Google AI Studio: Updated the icon to their new official logo—in case it looks different, that’s why!
  • Mistral: Fixed thinking steps in some models flooding your screen with
    [object Object]
    .
  • Mistral: Added support for the “Detect Characters” feature.
  • Updated the model list filtering in various providers to only show supported model types.
  • Removed the legacy “Anyscale Endpoints” connection.
🧑🏻‍💻 Prompting
While we continue to enhance our prompting system (and its documentation), we've cleaned up a few things and added/fixed some functions:
  • New
    min
    and
    max
    functions to get the minimum and maximum value of a list of numbers. For example: when using it in a custom shorten prompt, you can now ensure a min. word count.
  • Fix
    multiply
    not working as expecting - it now returns the full product of all provided numbers.
🔧 Other Changes
  • Fixed timeout-related loading issues with novels that exceed 100MB in total size.
  • Fixed some more rare AI streaming issues in Chat, which turned some messages blank.
  • Fixed placeholder labels in text editors not updating in some cases.
  • Various other fixes and improvements.