...
Browse GamesSteam news
Full Steam announcements and update articles for this title.
Hello Again, Readers, Writers, and Dreamers. A new Settings Menu is now in testing, designed to preserve all of our advanced features while making them far more approachable and user-friendly. Building on the settings overhaul introduced around Christmas, this updated interface maintains the same level of customization without overwhelming new users.
We’ve also introduced major backend upgrades for both text and image generation, adding support for the latest models. In addition, text-to-speech support has been integrated into our newest beta build and is currently undergoing testing. We're only going to include one voice when we launch the feature but we'll be adding alternate voices shortly afterward.
More to follow soon.
Infinitely grateful for your support. Chris
Happy Holidays to our wonderful customers! We've rolled out a big update to our main branch that adds the following:
Inference Engine Updated - We now support text models created before September 2025
Image Engine Updated - Latest Image backend has support for Custom Stable Diffusion 1.5, SDXL, and Flux models with custom VAE, Clip, T5, and LoRAs.
Revamped Settings Menu - Added options in the settings menu for browsing to your own model files, VAE, Clip, T5 and LoRAs along with a slider to choose between multiple customizable presets. You also have the ability to select which GPU is used for Image Generation for users with multiple GPUs (Previously everything defaulted to GPU0)
Cloud Support for Google Gemini API - You can now use google gemini based cloud models for text and image generation. More cloud providers to follow soon! *NOTE* Cloud Generation will send your story data and input to google, so if privacy is important to you we still recommend using local generation options. I think this is obvious but still thought I'd mention it.
A link on the image and text generation tabs in the settings menu to download new models from CivitAI (image models, VAE's, Clip's, T5's and LoRAs) and Huggingface (gguf format text models)
Fixed a bug where RTX 5090 GPUs were not being properly detected
As always, feel free to post on our discussion section if you encounter any issues or have any questions - I'm always here to help! Happy Holidays and thank you for your support!
We’re still hard at work — and while the new settings menu is still in beta, there’s a good reason: 🔧 We’ve been busy upgrading the llama-cpp backend DLL to support the latest models.
Why now? Because the new settings system makes it super easy to load your own custom models and we wanted to make sure the latest models work right out of the box. Maximum compatibility = maximum creative freedom.
✅ This updated backend DLL is now live in the Open Beta branch and ready for testing. Try it out with your favorite GGUF models and let us know how it goes! P.S. We’ve also included a new optional model file in this beta: LFM2, a brand-new release from Liquid.ai. It’s impressively lightweight at just 1.2GB, making it a great option for those running on 6GB GPUs or even 8GB GPUs with other apps running in the background. It also runs surprisingly fast with CPU based inference! If you’ve struggled with getting 4GB models like Phi-3.5 to run smoothly, LFM2 should offer a fast, responsive alternative that still delivers fairly solid results.
That said, due to its small size and limited context window, LFM2 may occasionally repeat page content or miss instruction cues. It’s only about three weeks old, so there's a good chance the creators are actively improving its instruction-following behavior. In the meantime, it’s still a fun model to experiment with while we continue to investigate these quirks and complete our rollout of full cloud support. Due to it's quirks it is not enabled or used by default, you will need to add it via the settings menu's Text-Generation tab.
Regardless of the model selected for general generation, exit choice generation will continue to use the default Phi-3.5 model for now, thanks to its strong and consistent instruction-following performance.
We’re also making great progress on a modular cloud provider system, which will allow us to plug in different cloud services for those of you who prefer SaaS-based (cloud service) storytelling.
Right now, we have support for:
🦙 Ollama
🧠 Gemini
More are coming over the next two weeks, and we’re planning another Open Beta update soon that adds this cloud support for everyone to try. This modular approach means easier maintenance, better performance, and more choices for you. Here's a sample video showing the speed you can expect when using Gemini to generate the story page, exits, and image: https://imgur.com/a/9QMEiZD Since all the 'heavy lifting' is done by a cloud provider, it should run this fast on almost any hardware setup.
Thanks again for being part of this journey! Your feedback and support make it all possible. If you haven’t tried the beta branch yet, now’s a great time to jump in and help us shape the next evolution of local (and soon cloud-powered) storytelling.
As always, happy adventuring Chris
I’ve got a lot to share - but if you're the tl;dr type, scroll to the bottom for the short version. When I started building this game, I had a vision: true local AI storytelling where your stories, your choices, and your data stay completely yours. No servers, no subscriptions, no compromises. Just you and the infinite possibilities of AI-powered storytelling - without getting cut off right when things are getting good because you hit a token limit. I've been there and it sucks.
But this journey started even simpler than that. I wrote the first prototype version of this for my 3-year-old daughter Molly - to make our storytelling nights more magical. Then I started having so much fun with it that I thought other folks might enjoy it too.
That's why we're here today.
I launched this on my late Grandfather's birthday toward the end of March, not knowing if anyone would care about local AI gaming. Some Steam games sell 20 copies and disappear. But y'all? Hundreds upon hundreds of you believed in this vision. You bought it, you played it, you gave feedback, you stuck with me through rough patches.
Thank you, truly 🥹.
Without you, this would just be a bedtime story app for Molly and I. Instead, it's become something real, something meaningful, something that's growing every week because you made it matter. Thank y'all for helping to make this better for everyone else.
Every sale, every review, every bug report - you turned a dad's coding project into an actual game that people love. That's humbling beyond words. I owe you and will continue to support this for as long as you guys keep showing up to play and purchase it. That's the least I can do in return for the trust you've given me. You trusted me with your money and your time. That trust means everything to me.
I spent the whole weekend rebuilding our settings system from the ground up. No more guessing what text models you have configured or image models you've copied over the standard files.
The new system gives you complete transparency and control over what is running:
🎚 Quality Sliders – Simple Low/Medium/High presets for both text and image models
🔍 Visual Model Paths – See exactly which model file each quality level uses
📁 Easy Browse Buttons – Click to browse and set custom models for each quality level
📦 File Size Display – Know exactly how big each model is before you use it
✅ Active Indicators – Clear checkmarks show which quality setting you're currently using
🧠 Smart Fallback – If a custom model goes missing, the system automatically reverts to using the included SD1.5 or Phi 3.5 model - no crashes, no broken text or image generation, no excuses
📂 Smart Organization – Everything grouped logically so you can find what you need
This isn't magic that reads your mind - it's transparency that respects your intelligence. You can see what's configured, easily change it, and understand exactly what the system is using at a glance. We've also added some much-needed enhancements to image generation. There's now a stop button that cuts things short when you've picked a model that's running too slow. And when generation finishes, a new regenerate button gives you another shot if the image doesn't match your vision. Keep going until the AI delivers exactly what you're picturing. The new settings system and image regeneration feature is live on the Open Beta branch on Steam right now. Please help me test it, break it - tell me what you like and what you don't. Together we'll polish it until it's perfect.
Many of you want storytelling magic but don't have the hardware for local AI. I hear you loud and clear. SaaS support is already in development - ChatGPT, Claude, and other cloud services integration is actively being worked on. Yes, you'll still hit their token limits and subscription walls, but at least you'll get to experience the storytelling magic while you save up for proper hardware.
Your vision matters as much as mine. I believe in local AI freedom, but I believe in your happiness even more.
Not only that, but we've also got a savegame manager in the pipeline right now too that will let you manage multiple saves and easily tell between them at a glance. This will allow you to have different stories, going at different times, to suit whatever mood you're feeling at the moment.
This is our journey, together. I've been trying to maintain weekly updates, but it's been two weeks since the last one because I've honestly been running low on steam (no pun intended!). Building this while juggling a day job and family life - coding this at 4:30 AM before my darling Molly wakes up - is taking its toll. But I keep pushing forward because you've shown me this matters to more than just me. Thank you all for giving me my 2nd wind.
There's one more way you can help shape this journey: reviews. Whether positive or negative, every piece of feedback helps me understand what's working and what needs improvement. Plus, it helps other storytellers discover what we're building together. I know leaving reviews isn't usually top of mind (it isn't for me either!), but it would mean a lot.
With my deepest respect and neverending gratitude, Chris P.S. - Molly, if you're reading this in 20 years: Daddy built this because he loved our story time so much he wanted to share that magic with everyone. Every line of code was written with you in mind, every feature designed to make storytelling as wonderful as our nights together. You were my inspiration then, and you always will be. tl;dr:🔧 NOW IN OPEN BETA: Complete settings overhaul with quality presets, easy browsing, and the control this technology deserves. COMING SOON: SaaS integration and a Save Manager for managing multiple savegames.
✅ Released: Flux DLC Model Pack The Flux model pack is now available as a free DLC! If you’ve been enjoying the enhanced image generation quality Flux provides, you can now re-enable it by downloading the optional Free DLC directly from the Steam store page. We originally removed Flux from the main install earlier this week to reduce download size and disk usage — especially for users who didn’t need the advanced models. This new setup keeps the base game lightweight while still offering power users full access to Flux’s high-quality image generation. 💾 How to Enable Flux:
Go to the game’s Steam page.
Scroll to the DLC section and check the box for Flux Models.
Once it finishes downloading, launch the game and head to Settings → Image Generation.
Toggle Flux on — and you’re good to go!
Alternatively: If you're viewing this post immediately after the DLC's release, the dlc may not yet be viewable on the game's main steam page until steam updates their caches. You should still be able to download it via search or via this direct link.
We’re still huge fans of Flux’s visual output and thrilled to bring it back in a way that’s more efficient and optional for everyone. Thanks again for all your support and input! With our Neverending Gratitude, — The Neverending AI Storybook Team
Hey Constant Readers,
We've just rolled out another huge update, focused on giving players more control, wider hardware support, and a better story-tracking experience.
Whether you're deep in Story Mode or just starting out, this update brings long-awaited features and key fixes -- and sets the stage for what's next.
We’ve added a new in-game tool to help you stay oriented during structured story mode:
✅ Chapter Tracking: You’ll now see how many objectives you've completed (e.g. "3/5 complete") at the bottom of the screen.
✅ Hover for Details: Hover over the chapter name to see a full breakdown of what's been completed and what's still in progress. These are hidden and only available on hover so that we don't spoil the story for you. If you've ever wondered why the AI keeps trying to send you back to the same place -- this will make that logic crystal clear.
✅ Flexible Progression: You don’t have to check off every objective to move forward -- completing at least 70% of them is enough to unlock the next chapter. You can follow your story, bend it, or complete it however you'd like.
Once you finish the final chapter, you’ll automatically transition into Free Exploration Mode, where you're free to explore, create, and continue the story indefinitely (or as far as your hardware will allow).
You can now save your progress and pick up where you left off -- even including your last generated image!
The current system supports one story at a time and includes your full story state — including character, item, story, and location memories -- chapter & beat progress, as well as your last generated image.
A full save manager (with multiple slots) is already in development and will be rolled out in a future update!
We've made big changes under the hood to help more players run the game -- even on lower-end hardware that is below our recommended requirements:
🧠 CPU Inference Mode
We’ve added options in Settings to switch both image and text generation to CPU-based inference by setting the model to use 0 GPU layers.
While CPU mode is significantly slower, it allowed us to successfully play the game on Intel integrated graphics (13900K) at a reasonable speed -- and we’re thrilled that this gives more players a shot at enjoying the game.
⚠️ Note: System monitors may still show GPU activity in CPU inference mode. This is normal - the underlying llama.cpp engine initializes GPU contexts and keeps memory 'warm' for offloading, but all actual text generation happens on your CPU.
📊 Smart GPU Detection + Manual Override
We detect your GPU by name and now match it to a built-in VRAM database to automatically decide how many LLM layers to offload.
And for tinkerers:
✅ New Override Field in Settings: You can now manually enter the number of layers to offload if you want finer control.
📉 If your system is tight on VRAM (especially if it’s shared with background apps), lowering the offload count can greatly reduce instability.
These tweaks mainly affect a very small number of systems — but if you're playing on tight VRAM, low-end hardware, or complex Windows environments, you may see better stability here.
We’ve also improved compatibility with systems that were just at or barely under our recommended spec or have uncommon configurations -- especially systems affected by aggressive VRAM limits and with different protections or antivirus software that may prevent writing to folders where we save the game and settings data.
Fixes include:
Display mode fallback logic (if your driver doesn’t support switching resolutions)
Safer file system handling (for environments with locked folders or aggressive antivirus)
More graceful failures in weird setups (instead of crashing or hanging)
We've removed support for Flux image generation models and are no longer bundling them with the game. If you were using Flux previously, the game will now automatically fall back to the standard Stable Diffusion model -- no action needed on your part. We will be adding them as a DLC package (free, of course) later this week, that way customers that don't intend on using them don't have to waste disk space on them anymore as they more than doubled the size of the game! We are still huge believers in flux as the quality of generation really is fantastic, we just want to ensure we deliver it in the best possible way for our customers that aren't running high powered battle-stations. More to come on this in a midweek update!
We’re currently testing a redesigned settings UI that’s a massive step up in usability and clarity. 📸 Here’s a preview of what’s coming:
We’ll share more about this in the next update -- but we wanted to keep you in the loop on what's coming next! With all our love and neverending imagination,
🪄 The Neverending AI Storybook Team
🚀 Weekend Update Incoming: StoryMode Progress Visibility & Save System!
Hey Constant Readers!
We've got some major improvements dropping this weekend that we're super excited to share with you!
📖 New StoryMode Objective Viewer System
You can now see the structure behind your adventures:
Visual Progress Tracking - See exactly where you are in your current story chapter with objective completion counters like "3/5 objectives complete"
Interactive Requirements Tracker - Hover over the current chapter title to see what objectives you still need to complete, so you'll never get lost wondering why the AI keeps guiding you back to "the elder's hut" or nudging you back to specific characters
Post-Story Freedom - Complete all Acts and Chapters to seamlessly transition into "Free Exploration Mode" for neverending adventures even in Story Mode
💾 Save System (Finally!)
The most requested feature is here! You can now save your story progress and pick up right where you left off. While this initial version lets you save one story at a time - we're planning a full save manager with multiple slots in a future update, but we wanted to get this essential feature in your hands ASAP.
🎯 What This Means For You (aka tl;dr)
No more lost progress or wondering "what should I do next?" The new system provides clear guidance while maintaining the creative freedom that makes Neverending AI Storybook special. Start an epic adventure, save it, and continue whenever inspiration strikes!
Rolling out this weekend - keep an eye out for the update notification!
Thanks for your continued support as we continue improving the ultimate interactive storytelling experience! 📚✨
With our sincerest neverending gratitude,
- The Neverending AI Storybook Team
About
This page aggregates Steam news feeds, patch notes, and developer announcements for Neverending AI Storybook, sourced from official Steam community posts.
Major updates, balance changes, and seasonal events often correlate with player-count spikes — cross-reference announcements here with the live charts on the main Neverending AI Storybook statistics page.
Articles link back to Steam for full changelogs. SteamScope refreshes news entries as new posts are published to the game's Steam hub.