Multiic

0.8.45: image generation · Supertonic 3 · video · 3D models · narrator · help image generation: prompts now adapt to the model: CLIP-based models get tags, while Qwen / Flux / Chroma / Anima and similar models get full sentences. The scene director also describes pose, camera and lighting in natural prose for these models. The prompt limit for them has been increased from 900 to 1800 characters, so scenes no longer get cut off mid-sentence the “Mielgo-like” style is no longer automatically mixed into local models ComfyUI generation timeout increased from 2 to 10 minutes, so heavier models have enough time to finish three new ComfyUI workflows: Qwen 2.1, Qwen 2.1 with a face detailer, and a faster detailer version. The multi-reference face generation workflow has also been improved first-person mode. If you disable “describe yourself”, you become the camera: only the characters appear in the image, and your face is never shown. When you physically take part in a scene, only your hands enter the frame from below (and during sex, your body as well). Kisses are rendered as “the character leaning toward you” instead of two faces pressed together. Other first-person camera angles were also changed — instead of a distant wide shot, you may get a top-down view through your own eyes, with the character shown full-body new Supertonic 3 speech synthesis runs entirely on CPU — no GPU or server required supports every language in the game except Chinese breathing, laughter and sighs are performed by the voice itself video: fixed: transitions no longer interfere with generated video playback fixed: animated backgrounds no longer flicker or freeze because of the cursor effect, and the menu no longer gets hidden behind generated videos fixed: the “frames” slider now correctly displays the saved value videos are converted to WebM instead of heavy GIF files added ComfyUI workflows for Wan 2.1 text-to-video video timeout increased to 30 minutes, with clearer ComfyUI errors generated videos can now be saved as WebM 3D models: models for Laura, Dina, Jo, Mika and Tisha have been completely reworked modular clothing in 3D: outfits are now equipped on top of the character without reloading the model; added a new “clothing” button to the menu. Default outfits have been assigned to the characters added an extra hairstyle for Laura and glasses for Tisha lip sync completely rewritten conversation animations reworked fixed bugs when switching models quickly lighting now adapts to the background improved cursor gaze tracking and smoother return of the character’s gaze back to the player camera zoom during speech is now smoother updated the set of six idle animations and removed the weaker ones fixed idle animations resuming after speech and other animations added smooth turning toward the player, which can be interrupted by manually rotating the camera narrator: no longer confuses you or the current characters with the full character list if a character has a character card assigned, the narrator uses the name from that card narrator lines no longer have a nameplate relationship status checking relationship status no longer interrupts the character’s response displayed on a separate line: no name, centered, green now also appears when streaming is disabled (previously it did not) the character card name is now used instead of the default character name character cards fixed: even if all base contexts are disabled on a card, the character still receives the rule that everyone is over 18 fixed the “do not limit AI response length” option (previously it did not apply when triggers were disabled for the character) fixed: character cards now work properly in Tisha’s slot with any context configuration. Previously, disabling certain checkboxes could make her lose conversation history and part of the context support for GPT-6 and newer models. If the provider responds that a parameter is unsupported (temperature, top_p, etc.), the game removes it and automatically retries the request updated Whisper speech recognition gallery: 14 new images — Mika +4, Tisha +3, Kas +3, misc +4 KoboldCpp updated to 1.121 crypto payment button hidden (NOWPayments disabled payments for me because of my passport. Nobody was using it anyway) help screen redesigned: cards, icons and a quick-start section; new FAQ section with 12 common questions about APIs, local models, image generation and saves; recommended model list updated tooltips rewritten clicking the new version number on the main screen now opens the devlog page on the website reworked the “character starts the conversation” option: opening scenes will no longer always be the same; the character remembers where you left off; if you have already talked before, they continue the conversation and take into account how much time has passed. This works even with conversation history saving disabled — the character still remembers the last few messages. First meetings are always varied, and recent variants are not repeated. Character cards now use their own greetings when one is provided in the card Saiga Nemo is back as the default text model 3D: Jo got the highest-quality model so far (still not without flaws). It is still quite far from the canonical references, but now I know for sure that getting there is possible I’ll keep improving all the characters in the same way, including the male ones. let me know if you need voice cloning for Supertonic — I’ll upload it separately --------- 0.8.19: Tisha, story, french, german, spanish, chinese, ukrainian, portuguese (br), jinja, transition effects, local model rebuild, qwen3.5-9b-defiant-fable-uncnr-heretic-neo-max-mtp, bug fixes. gallery. image generation → Tisha’s story event has been completely reworked; u can find it in story navigation under “therapy session (Tisha)” → Tisha is now a full sandbox character (woohoo); you can also visit her for a therapy session in sandbox mode → use jinja - a new toggle in settings. launches koboldcpp with the --jinja flag, so utility requests (image prompts, clothing, detectors, etc.) use the model’s native template → added the ability to generate up to 3 images from a single prompt; the setting has been moved to the “image prompts” section → the game is now fully translated into french, german, spanish, chinese, ukrainian, and portuguese (br) → 4 sweet new transition effects for background changes and generated images → the default text model has been changed to Qwen3.5-9B-The-Defiant-Fable-Uncnr-Heretic-NEO-MAX-MTP-Q4_K_M.gguf → “continue” story button on the main screen → the hyperwoo minigame has been updated to 0.1.35; completing each level now rewards 20–30⦵ → Mika (180⦵) and Tisha (135⦵) can now be unlocked in sandbox before they appear in the story → new gallery items for Tisha, Finn, Kas, Fiona, Laura, and Dina → reworked some of Mika’s weaker story images from the first encounter → fixed image prompts (and other small utility bots) for certain models, especially gemma → server-lv, previously used to bypass blocks, has been retired (due to sanctions, beget’s latvian server stopped working); all of its functionality has been moved to server-kz → changed ua to uk for language AI, in accordance with the ISO 639-1 standard → claude-sonnet-5 is now selected correctly from the recommended models → fixed several story dead ends → redesigned the main screen and context settings. language dropdowns have also been standardized → koboldcpp updated to 1.119 → added rtx 50-series gpu support for local speech synthesis/recognition and stable diffusion; older gtx 10-series cards (1050–1080) are no longer supported and will fall back to cpu. the same will happen if yr nvidia driver is outdated - version 580 or newer is required → coqui: torch cu121 → cu130; fixed voice cloning - the reference recording no longer falls back to a preset voice; the voice list also no longer requires restarting the game if u switch to coqui while the game is running → qwen-tts: torch cu118 → cu130 → updated the coqui and qwen links in settings and help → whisper and stable diffusion rebuilt with cuda 13.3 → stable diffusion: updated the engine; added support for wan, ltx, qwen-Image, z-Image, chroma, and new samplers; “hot model” now works differently - when enabled, the model stays loaded in vram; when disabled, it stays in system ram, leaving vram free for other tasks; the build is also smaller now; image generation on nvidia gpu has been fixed on linux → code refactoring → local text models: fixed reasoning, system instructions, and service tokens like user / assistant appearing in character replies improved compatibility with Qwen, Mistral, Gemma, and other GGUF model families both regular and streaming responses are now more stable fixed connection to remote KoboldCpp instances bypass eos is now disabled by default; cases where it is actually needed have been reduced to a minimum KoboldAI has been deprecated — it hasn't received updates for two years, and its own developers recommend using KoboldCpp instead → image generation models: your self-description now reaches the model completely untouched more room for descriptions: “describe yourself” — 180 instead of 90, character appearance in character cards — 450 instead of 270 prompts are no longer blindly truncated. the limit now depends on the model, and if space still runs out, styling gets trimmed first instead of the actual scene: pose, action, and camera angle are never discarded scene generation is now more accurate. the model separately determines who is in the frame, what is happening, the shot type, and the lighting — instead of processing it as one continuous line fixed counting people in the frame: extra people should no longer appear, and required characters should no longer disappear character appearances are preserved better between characters, with less feature mixing and fewer traits from one character leaking onto another. distinctive traits of characters who are not in the frame are now automatically excluded the scene location is no longer replaced by the conversation context fixed cases where, after a failed model response, service text could end up in the image prompt instead of the actual scene description you no longer need to specify your age or gender in “describe yourself” — they’re taken from your date of birth and the main screen settings. if the field already contains your age/gender, duplicates are automatically removed from the final prompt new “don’t use” toggle next to the “describe yourself” field: excludes your appearance from generated images and fades out the field. useful if your features start bleeding into the character when generating multiple images at once: the first shot is framed however the “director” intended, the second is a close-up, and the third is a wide shot. previously, images in a series mostly differed only in noise/grain and ended up looking very similar; now they’re different shots of the same scene the tooltip for the “describe yourself” field has been rewritten and now explains all of the above comfyui: added support for PrimitiveStringMultiline / PrimitiveInt nodes for Krea Flux models → fixed the layout of the API models section on the settings page → added a progress bar when unlocking/locking all gallery items → changed Tisha’s starting locations in Sandbox → fixed location synchronization in Sandbox between the text and image models, including the office location → reworked the way characters are called into a scene from dialogue: this is now handled by a separate bot that knows character card names, understands exactly who was called, and ignores vague requests; duplicated dialogue lines when a character arrives are also gone ! bypass eos is no longer a recommended option. it’s better to disable it, since local models now correctly understand when they should stop generating. with it enabled , the model will keep generating until it reaches the response limit the new text model replacing Saiga Nemo has fewer parameters (9B vs 12B), but in practice it’s smarter, faster, and better tuned for natural roleplay dialogue generating up to 3 images from a single prompt will be especially useful with local models at a slower game pace, letting each moment play out more fully and making playthroughs feel more varied gtx 10-series support: the rtx 50 series requires cuda 13, which no longer supports the Pascal architecture. maintaining two separate builds for different GPU generations would mean several extra gigabytes. according to current Steam statistics, only around 1–2% of users still have these cards, and local models are too slow on them for comfortable gameplay anyway. so, sorry, but I decided to drop them from Stable Diffusion and the local speech models. as an alternative, koboldcpp supports every GPU generation, so use that if this affects u --------- 0.7.68: custom contexts and character cards. hyperwoo. one thousand and one nights. Mika. story → reworked custom contexts → png/json character cards and loras for custom characters → ava is no longer ava — her new name is Mika → story: - the prison section has been completed and connected to the rest of the story; - a new d&d hospital section has been added, with a new “hospital (one thousand and one nights)” point in story navigation; - playing hyperwoo with Mika: the choice about humanity from level 6 of the mini-game is now saved to Mika’s context. she will offer to play it in prison and when you choose her at night, after the park scene with Jo, Fiona, Dina, or Finn → a new one thousand and one nights d&d scenario in sandbox mode → tamagotchi: added grok voice support and a voice selector in the tray menu → new music → koboldcpp: contextshift/swa settings are now available directly in the game →