Hi everyone!
This is a bit of an experiment. I wanted to see if it was possible to create something decent using only open-source tools. I know it's not perfect, there are a few small glitches and inconsistencies, but I’d say the result is pretty good for a first attempt.
I mainly used free and open-source software, all stuff you can easily find right here on Civitai, Hugging Face, or GitHub. The only paid tool was the video editor I used to put everything together.
Here is the tech stack I used:
Images: Qwen 2512 text to image to generate the images, anything2real (a Qwen LoRA) to enhance some of them, and Qwen-edit 2509/2511 for editing them.
Video: LTX Video (v2.3) base model with the original workflow.
Voices: ChatterBox.
SFX & Music: Pixabay.
Script: Gemini (for translating, refining dialogues, and brainstorming).
If you want to see how the story continues and support my work, please leave a like and share!
I noticed YouTube significantly degraded the video quality after the conversion. If you want to see the footage exactly as it was meant to be, with proper details and no compression artifacts, check it out here on Civitai: Lineage