Skip to main content

Browse GamesGame JamsUpload GameDeveloper LogsCommunity
Indie game storeFree gamesFun gamesHorror games
Game developmentAssetsComics
SalesBundles
Jobs
TagsGame Engines
andevol22 days ago

Hi, I thought I would try this before purchasing so I have installed the light version.  I am having some issues with chat generation, and submit the following for your information.  I created a new profile with no uploaded image, here's an image of the results, please let me know if you want the information produced in the Dating App Server console:

Bug Report: Dual-output JSON Parse Failure Symptom: Chat responses display as static placeholder images instead of text Console Log Evidence: - LLM generates 215-220 tokens successfully (confirmed via timing logs) - HTTP 200 OK responses from llama-server - WARNING:llm_client:Dual-output JSON parse failed, raw='' - Image appears that is NOT from Stable Diffusion (not SD-generating) System Specs: - RTX 3060 12GB - Windows 10/11 - Using llama-server.exe + Gemma-4 GGUF model - Tested with SD disabled (still shows images) Expected Behavior: - Text response from LLM should display in chat - Images should only appear when explicitly requested or SD is enabled Actual Behavior: - All responses render as hardcoded placeholder image - Text from LLM is discarded after successful generation Likely Cause: - Heartsync expects dual-output JSON format from LLM - Generic llama-server only returns plain text - Parse failure triggers fallback to placeholder image

Reply
Inoxannin21 days ago

Hi andevol, thanks for the detailed report — really appreciate the console log evidence and the structured write-up, that made this easy to track down.

Found it: you had a reasoning model selected, which "thinks" internally before producing its actual answer. That internal reasoning was eating up the whole token budget, so nothing was left for the actual response — the dual-output parser then received an empty string and fell back to the placeholder image. For this app I'd generally recommend an instruct model over a reasoning model — reasoning models also tend to take noticeably longer to generate, on top of this issue. That's fixed now and already live in the current upload, so reasoning models should work correctly too if you prefer to stick with Gemma.

On the image trigger issue: you're right, that one's still open. Right now the app detects image intent based on a simple keyword match, so any message containing the word "image" triggers generation, even in something like "don't generate an image." I haven't fixed that one yet, but it's on my list — I'll follow up here once it's resolved.

Feel free to grab the current update and test the reasoning-model fix in the meantime. Thanks again for taking the time to write this up so thoroughly before even purchasing.

Happy dating!

Reply
andevol19 days ago

Hi Inoxannin,

Thanks so much for your reply and for looking into this.  I will test the reasoning model, but also try an instruct model.  

Reply
AboutFAQBlogContact us
Copyright © 2026 itch corp · Directory · Terms · Privacy · Cookies