Multiic на андроид

0.4.27: - 2025-03-22
the plot
, animation on the start screen
✨ micro-animation and sound of the clock when some time has passed
✨ some pictures for Joe
✨added gemini-2.0-flash-lite model, eliminated gemini-1.5-pro-exp-0827, gemini-1.5-flash-latest and gemini-1.5-pro-latest
🛠 koboldcpp version has been updated to 1.86 (supports gemma 3)
🛠 Fixed a bug where saves and settings disappeared within 5 seconds when closing the game and reopening it.
When using local koboldcpp CR models, it is still recommended to wait 2-3 seconds before reopening the game.
When importing the settings and save file , information was added that a reboot of the game is required .
The music in the storyline has been corrected
🛠 language change and no longer requires restarting the game
The sandbox setting for the character to speak first no longer requires restarting the game.

0.4.12: contextual field, kisses, statistics, claude, gpt-4.5

→ new context field for describing the world and instructions
→ micro animations and sounds during a kiss
→ a few new park locations
→ added claude-3-7 model
→ added gpt-4.5-preview model
→ updated section with lore and characters, statistics added
→ apk: fixed claude's functionality
→ optimized character system contexts in the sandbox and images in some scenes
→ now Laura is 28, and Finn is 29 years old

statistics: for now it just counts how many times the character and the user cum -`♡´- will be further developed so that it can be used in the system context

0.4.05: gallery, mini-game, bugfix

→ gallery: added a multitude of images, now to open them you need to use ⦵, which can be farmed by winning mini-games: chess, save the princess and collecting macrodata
→ Jo, Laura, Fiona and Ava: now they can go anal (trigger and images)
→ Laura and Jo masturbate (trigger and images)
→ a gallery option has been added to the quick menu
→ mini-game with macrodata (the very same ones collected by the workers from the show - severance), available from the phone or by asking Ava
→ added model gemini-2.0-flash
→ groq: added model llama-3.3-70b-versatile, deprecated llama-3.2-90b-vision-preview
→ added park locations
→ fixed bug with english context when adding a second character
→ koboldcpp updated to version 1.84.2
→ removed the button for activating 3d from the story section
→ apk: fixed voice selection for elevenlabs
→ fixed bug with the car, when the location did not change after the car
→ person and base passwords changed

for "save the princess" the most points are given - 40
for chess (losing is almost impossible) - 10
macrodata - random from 1 to 18

macrodаta: the idea was spotted and swiped from the expanses of codepen
for everyone who hasn't watched the series (severance) - I highly recommend it

0.3.97: speech elevenlabs, 3d lipsync demo
→ speech synthesis from elevenlabs (also on apk)
→ sandbox: demo of 3d characters with lipsync
→ when first opening the game, the language selection for ai is now set correctly
→ selection of cosmosrp-pro and chatgpt-4/4o models is open and no longer requires a game subscription (without person password)
→ updated list of starting models on the welcome screen and in settings (if there is no local model in the game folder)

3d:
one of those ideas that came almost immediately but was put on the back burner my knowledge in 3d is quite superficial and not enough to create a model myself, so I used the free very limited editor ready player me, models are boring without any hint of sexuality, but together with lipsync demonstrate what I want to do

plan: find a suitable 3d artist on the marketplace and order a character based on refs from the game (later we will decide who will be first)

approximate level of the model that I want

if u know something that can help - write ) the model should be compatible with ready player me + arkit, oculus blend shapes

with the 3d avatar, only speech synthesis from google and elevenlabs works so far,
fps locked at 45 frames

speech synthesis from elevenlabs in my brief view is top: a large selection of voices, trial, slightly cheaper than openai, and no worse in terms of level

the continuation of the story is still in development. unfortunately, plans have shifted somewhat (there were many important bug fixes). the new deadline for continuing the plot is the end of february to the beginning of march

0.3.92: bugfix, locations, eos
→ sandbox: when changing locations from the menu, characters comment on the movement. you can set a favorite/start location for each character
→ added bypass_eos setting
→ updated koboldcpp to version 1.82.4
→ updated help section (hoping it’s more user-friendly for new users)
→ fixed auto-save: the slider now also affects automatic dialog trimming. the recommended value is set to 18, and the maximum reduced to 40
→ fixed steam activation in the shower when selecting locations from the menu
→ fixed AI language selection in dark theme
→ winter theme extended until February 13
→ fixed "rescue the princess" minigame (my crooked hands forgot to close a parenthesis in the code, damn it)
→ sandbox response options setting now works correctly with generated answer variants (it was broken and worked the opposite way, hehe)
→ story: now, at the end of a short series dedicated to the communication part, if a character asks a question, the answer is not replaced with an ellipsis but is fully displayed with the note "the question remains hanging in the air."
→ koboldcpp logs are now always trimmed to the last 360 lines
→ when opening the sandbox, the "what happened before" button now correctly appears only if the character's history was saved
→ fixed the order of save/load slots
→ fixed dice flying off the table in DnD: now there is an option to re-roll

bypass_eos - use this if the local model outputs "EOS token triggered" too frequently and inappropriately

there was a bug with auto-save - a hardcoded value of 20, which when a higher value was set in settings (now also affects trimming), was simply ignored and saving did not work

0.3.86: phone, ai language, bugfix, mac
→ phone call mechanics in the sandbox. it’s for the storyline part, and in the sandbox it’s used for testing, adds a bit more immersion
→ added ai language selection (Français, Español, Português, Deutsch, Italiano, Українська, Slovenský, العربية, Türkçe, 한국어, ภาษาไทย, Polski, 日本語, Tiếng Việt, ქართული, Հայերեն)
→ fixed the die for dnd modes to a d20, added sound, and made some visual adjustments to the buttons
→ if a local model is selected, the game no longer requires a network connection
→ fixed version checks and password checks when offline; in some cases—especially on smartphones—they could cause the app to freeze
→ person and base passwords changed

ai language: this only affects the text from chatbots. because the system context is written in english, sometimes the characters may slip into english, but in the sandbox this happens much less often, and it’s easier to get them back into the chosen language

the phone mechanic is new and may be buggy on some models

mac: for now, without support for the local model—I’ll try to integrate it next week
on my friend’s macbook, everything launched without additional permissions, but it might not be the same for u.
the prblem: apple requires the app to be signed, and it costs an additional $100 per year for that (greedy). without a signature, u might need to grant extra permissions in the settings to run the application

0.3.80: local model + koboldcpp
0.3.80: Local model + koboldcpp

koboldcpp is integrated with the local model*
✨ desktop: added a welcome screen with a choice of local model or API
✨ added location selection in the sandbox (upper left corner)
Now if koboldcpp and koboldcpp-remote are selected, the kobold tab is open in the default settings
✨ Ava 's failed images have been updated
Redesign of the selection buttons
✨ Redesigned the path to one of the sex scenes with Dina in the plot
The minimum screen size in windowed mode has been increased from 640x360 to 900x500
improved bots that detect the change of locations and the arrival of the second character. fixed operation on local koboldcpp CR models
🛠 The maximum temperature value has been lowered to 1.4, recommended to 0.72
🛠 koboldcpp:
→ now, when the game is closed, the koboldcpp and electron processes terminate correctly and do not remain in the task manager.
→ fix for early completion of token generation, (EOS token triggered error! ID:2)
→ adjusted settings
→ fixed the operation of suggestions and generated options
→ the maximum response length has been reduced to 630 tokens, the recommended value is 190
→ The koboldcpp shutdown button has appeared
→ added information about approximate system requirements on the start screen and in the settings (if there are no models in the folder)
→ added your own stop triggers
fixed a bug with version verification
🛠 the animation of the earthquake has been corrected in the plot (and possibly some more)
🛠 the music in the story event with Fiona in the shower has been replaced

*now there will be three options for windows: standard (with koboldcpp), -local (with a local model) and -lite (without koboldcpp), for linux you will need to download the model separately and select it in the settings (later I will figure out the configuration and do the same as on windows with the resources/koboldcpp folder

windows: koboldcpp with models is located in the resources/koboldcpp folder, all gguf models from this folder will be uploaded by the game to the start screen and tab in the settings

windows/linux: A model from any folder can be downloaded from the settings tab by clicking on "select model"
the local model I've settled on is Gemma-2-Ataraxy-v4d-9B.i1-Q4_K_M - it feels like ~ chatgpt-3.5, weak, but it handles emoji triggers much better than cosmosrp, I think it should work normally for most users

of the models that I tested and which is probably more suitable for the game, but more demanding - ChatWaifu_v1.4-Q4_K_M-GGUF (if there are no models in the resources/koboldcpp folder, it will show the system requirements and a link on the main screen and in the settings)

0.3.67: festive + bugfix

→ spirit of christmas and new year
→ grok-2 models added
→ now Dina, Jo, Finn, Laura, and Ellie can dress warmly if it's cold
→ winter city and forest locations added
→ functionality for transferring settings and game data added (bottom right corner)
→ Fiona: updated unsuccessful images, added new ones
→ snow added in locations where it can snow
→ for optimization purposes, a setting for showing images on loading and save screens added, now they are hidden by default
→ interface animations revamped
→ auto-save slots order fixed
→ dark theme fixed
→ bugs with {{usermessage}} and empty messages fixed (but not for sure)

data/settings transfer - a feature needed in the future, as well as for those who simply want to transfer from one device to another. it will be needed for android users a bit later when i fix the incorrect application identifier value

displaying images on loading/save screens - the problem is that if there are many slots, it can significantly affect performance on weak machines because game images are used instead of thumbnails, especially if they are animated backgrounds. i'll handle creating thumbnails later, for now, it's like this

0.3.56: d&d scenarios, cyberpunk, zombie, date

→ now all sandbox characters know what day it is today
→ added user’s date of birth selection at the beginning of a new game (can be changed in settings)
→ added setting to disable generated dialogue options in the sandbox
→ person: d&d scenarios added to the sandbox: cyberpunk 203x and zombie apocalypse
→ threesome: fixed the determination of the connectable character when called from chat

0.3.51: threesome, grok, bugfix

→ a feature with two characters on the screen added to the sandbox
→ support for the grok-beta model (the one from E Musk)
→ a button for forcibly returning Dina to the screen
→ a critical bug with custom contexts for the female hero fixed, now everything works including dnd mode
→ maximum temperature reduced to 1.8
→ gemini-1.0-pro and gpt-3.5-turbo-1106 models removed
→ dark theme fixed
→ apk: less screen dimming while typing
→ gemini: optimized the work of bots responsible for changing locations and saving history; the location-bot now works correctly, always creates a new empty session and only looks at the last 2 messages, not the entire history; the history-saving bot no longer has redundant duplicated text. especially relevant for long gaming sessions
→ in the prologue dream, the embarrassing (poor-quality) videos with dina and lora have been removed, replaced with images
→ music files re-encoded into ogg (same quality but half the size)
→ all webm videos slightly compressed without visible quality loss
→ person and base passwords changed
→ optimized/removed/replaced unsuccessful animations and images
→ fixed a bug with pairs, where reactivating the trigger added another pair, which could significantly reduce performance

threesome invocation in the top-left corner, can be called by text as well, but the menu is more reliable. it works like this: if two characters are on the screen, a hidden bot decides before each response who should speak, switching from one character to the other along with all their custom contexts

grok - interesting, has a free $25 trial, seems not to censor at all; be cautious with temperature, above 1 it starts hallucinating; gets buggy, repeats phrases in prolonged conversations

threesome issues with cosmosrp and gemini:
with cosmosrp and some local models, it may work poorly or not at all. the issue: the bot that determines who will speak should only return the name, and most models handle this successfully 100% of the time, but cosmosrp-pro for reasons unknown to me returns a long sentence completely unrelated to the request instead of the name. I complicated the logic to extract the name from the response if at least a name is present, but the name appears in less than half the cases, overall it's unmanageable, unfortunately :`(
to somewhat mitigate the shortcomings of such models – when there are two characters on screen – the speaking turn is given to one of them randomly 50/50
with gemini it's still a bit more complicated, due to its session features each character does not see the message history of the other, i.e., for him/her only what the user said is available, but not what the second character said. I don't know what to do about this yet, theoretically you could inject part of the dialogue into the system context, but it's not certain and I'm not sure it would work well

0.3.41: koboldcpp remote
→ koboldcpp: added support for remote mode; to fix the issue where characters start speaking for the user, added a field to exclude the name form - relevant for languages other than english
→ now api keys and password are carefully hidden
→ fin: 2 new images
→ redesign of settings page
→ help section updated
→ fixed the mechanics for selecting dialog/action options, should be a bit more reasonable; number of options increased by one
→ apk: fixed a bug where the dark transparent background of the overlay when minimizing the keyboard did not fully cover the scene
→ redrawn unsuccessful images in the hospital

0.3.34: Laura, changing locations in the plot, options
plot part:
→ new mechanic with generated dialog/action options*
→ now location changes also work in the plot part (brought more freedom, but possibly a bit more chaos)
→ redesign of pop-up windows
→ redrawn unsuccessful images of Dina, Fiona, and Laura
→ redesign of the event in the prologue at the table, fixed errors during Laura's communication with Dina
→ new event for Laura in the bathroom, in the prologue with doors.
→ person, plot navigation: became available not after completion, but immediately; added items with prologue (dinner and doors); fixed a bug when clicking on a navigation item, chatbots stopped working
→ apk: fixed display of the keyhole
→ a few new images for Laura in sandbox
→ background animations have been optimized, it should perform a bit better on weaker machines
*works ~ like suggestions, but: without the ability to write your own text, from 4 to 6 options, with attempts at humor and oddity.
also added to sandbox with a very small probability of occurrence <1% (more precisely, every 16th message has an 18% probability)

0.3.27: Ava, music, gemini, story fix, bugfix
→ Ava: added new images and starting locations; when connecting to user’s chip – cool scripted background
→ gemini: added models gemini-1.5-pro-002 and gemini-1.5-flash-002
→ 4 new tracks in sandbox
→ now information about new version is shown on main screen (if current version is outdated)
→ added keyboard typing sound (used in story part)
→ gemini: fixed bug where entered text on failed request was not added back to the field
→ story part: fixed incorrect display of pronouns and addresses for female player character; added female hero avatar in dialogs; fixed errors (in some places narrator was not loaded); fixed transitions and dialog errors
→ slightly optimized game’s initial loading

0.3.20: locations, kobold, d20, dnd-mode, sandbox saves
- new locations: stadium, green field, abandoned factory/warehouse, mage's tower/ancient observatory, ruins of the sky city, crystal cave, icy plateau in the mountains, bdsm room, rooftop
- kobold is still in beta but is now available to everyone without a subscription
- Dina: redrawn unsuccessful images, added new ones; fixed dina's return in case she left
- Jo: redrawn unsuccessful images
- person: saves in the sandbox
- dnd mode: rewritten system context for dnd mode; added a cool 20-sided die, now used for all rolls
- reworked auto-save history functionality every n messages. now it's an optional feature, disabled by default*
- slots are now displayed from left to right on the load/save screen
- tweaks to the start screen
- fixed issues related to incorrect display of the "Where were we?" button and history clearing when a part of the context is deleted
- changed passwords for person and base

0.3.13: temperature, chess, koboldcpp, streaming
- u can play chess with Dina*
- temperature (degree of freedom): maximum value raised to 2, recommended up to 0.81. values above 1.0 for quirkiness and greater creativity
- windows, linux: added a new cursor, less resource-intensive, should work smoothly on weak machines (if there are no complaints, i'll set it as default on first launch)
- sounds: dice rolling, shattered glass jar
- koboldcpp (beta): added streaming output, fixed system context passing and chat mode operation
- ux-ui tweaks
- minor redesign of the start screen on smartphones
- apk: fixed bug with overlapping tooltips
*for now, Dina doesn't comment on game moves, she just reacts to a win, loss, or draw. and it's very hard to lose when playing with her. if the idea catches on, i'll add control and game move commentary to Dina so she plays herself (currently, it's done by a rather dumb algorithm)

0.2.98 - 2024-10-26
0.2.98: koboldai (beta), cohere
✨ support for local koboldai CRZ models in beta status
✨ added models from cohere*
✨ filter by recommended models
✨ Claude models added-3-5- sonnet-20241022 and llama-3.2-90b-vision-preview
✨ en: the choice of the server for the internal proxy of openai and anthropic is made in the settings**

*at first glance, the gemini-flash level is not very powerful, fast, and expensive. they give you a free trial, which is nice
**the Kazakh server suddenly stopped working, or rather openai and anthropic stopped working in Kazakhstan, if I understand correctly. while I'm deciding where to move, I decided to throw it on the proxy where gemini is spinning, there the server is running in one stream, it may be more loaded than usual for some time

0.2.91: music, nvl mode, dark mode, fonts, cosmosrp, bugfix
✨ Added animation and sound when changing character images
✨ New music on the home screen and in the sandboxes*
✨ NVL-text display mode
✨ dark mode, dark theme added
✨ now the text input field has become multi - line
✨ font selection in the settings
🛠 fixed streaming output for gemini (without internal proxy)
🛠 fixed input field
apk: fixed dialog output in adv mode
🛠 fixed speech synthesis from Google
🛠 fixed the minimum size of the adaptive font
🛠 the button to skip dialogues is removed from the sandbox as unnecessary
Now the text from the magazine can be selected and copied
🛠 fixed the count of the number of characters in custom contexts

*so far, the god of random rules in music, random on the composition of the start screen and random in the sandbox. There will be playlists and track selections later.

v.0.2.76 - 2024-10-11
- Laura: updated images, character
- now the text is displayed with paragraphs, woohoo! *
- now in the sandbox when activating undressing triggers, the character's image updates automatically (a random one from the set is chosen)
- new images for Dina, Jo, and Finn
- each character has their own set of moans, person: can be reassigned in context settings
- Finn received his own branch in the prologue (first meeting in the event with doors)
- fixed buttons and triggers for skipping sexualized scenes, added sound effect **
- sandbox: the back button now inserts previously sent text into the input field, works only one step back. fixed a bug where changing the character's skin would reset the existing location/generated background to the default one
- gemini: when a limit or any other error is reached, the previously entered text will be carefully added to the input field for editing or resending
- fixed a bug where an incorrect endpoint for claude models was set when opening the game
- slightly modified parallax
- sandbox: fixed a bug with characters leaving, only dina can leave now, but she can always be returned
- plot: fixed an error with the old man when trying to pretend you know him
- fixed the translation of hints (3 quick answers)
- apk: now when clicking on the main game screen with the character, a smooth view shift animation is used
- changed passwords for person and base
*my joy might seem strange for such a small thing (actually, not so small) - this is one of those tasks you keep postponing, thinking it'll take several days, and then you’re surprised why it wasn't the first idea. Turns out it’s solved with just one line, the hero of this update - white-space: pre-wrap
**false triggers also fixed. important note: when using the free version of cosmosrp - due to model limitations, triggers for when the user expresses the desire to leave or the character wants to end the dialogue themselves may not work correctly. in such cases, the button will definitely skip the dialogue

0.2.63: hotfix and Dina
- fixed an annoying critical bug related to the protagonist's gender in the storyline. it now updates correctly based on the selection made on the main screen
- increased the character size
- corrected minor translation errors
- for Dina in the sandbox: slightly edited her personality, updated images, now capable of undressing down to her bra, and can leave if she’s offended or for any other reason*
- Fiona, Dina, and Ava now have 'come' imgs
*if the departure mechanics are functioning properly, this feature will be added to other characters

0.2.58: female protagonist, prologue rework, bugfix, Jo
- reworked prologue:
-- added the ability to play the story part as a female protagonist. (Fin will be woven into the prologue story in the next update);
-- replaced unsuccessful images;
-- redesigned user input windows in critical places to make transitions seamless;
-- enhanced some events;
-- added ambient sounds;
- apk: added the ability to copy text from the dialogue log (temporary solution, downloading the file like on desktop will come later)
- a few new images for joe in the sandbox
- attempted to fix empty messages when using gemini-1.5-pro models
- fixed narrator censorship issues
- improved location change detection in sandboxes for the eng lang
- now in all sexualized story segments, in addition to the special button to continue the story, you can finish just by indicating it in the dialogue, and characters can also decide to finish on their own
- fixed translation bugs
in the next update, I will finish reworking the story for a female protagonist

0.2.51
- revamped the poem event: fixed a critical bug that prevented progression; nerfed the required points from 27 to 23. corrected the display of the second attempt. now the event is not mandatory to complete and after 3 attempts the story continues, added an image in the gallery as a reward for completion
- due to performance issues, the boring cursor was restored by default
- fixed translation errors
- fixed errors with proxyapi and proxy, on desktop fixed the claude+proxyapi error

0.2.49
Initial Release