18 pointsby graphpilled10 hours ago4 comments
  • QuantumNomad_9 minutes ago
    Probably neat, but “Show HN” should be removed from the title as this is a signup page for a product you have to pay to use.

    > Off topic: […], sign-up pages, […]. Those can't be tried out, so can't be Show HNs. Make a regular submission instead.

    From the rules for Show HN posts.

    https://news.ycombinator.com/showhn.html

  • notsylver30 minutes ago
    This really needs some examples or a demo that doesn't require signing in. As-is this is just a pricing page with a sign in button and no information on what it actually does?
    • graphpilled21 minutes ago
      Good point! Here's a quick demo showing how it works: https://www.youtube.com/watch?v=b66ZHznBvEM
      • rusty__15 minutes ago
        Is that issue not immediately apparent to you? Your splash page says literally nothing. Why can't you put even a tiny example of what this is all about? You can't expect people to hand over personal information to hear an elevator pitch.
  • avaer21 minutes ago
    Seems cool, but is there more to this than just some outlines in a graph? That's all I was able to try without paying.

    You mention there is "fountain exports" so I was expecting something more akin to a live-morphing screenplay kind of thing (and I would be the ideal customer for that, I work on procedural roleplaying video games).

  • graphpilled9 hours ago
    Hi HN, I'm a computer systems engineering student in Mexico who switched from film school. I built CineGraphs because my filmmaker friends and I kept hitting the same wall—we'd have a vague idea for a film but no structured way to explore where it could go. Every AI writing tool we tried output generic, formulaic slop. I didn't want to build another ChatGPT wrapper, so I went a different route. The idea is simple: you input a rough concept, and the tool generates branching narrative paths visualized as a graph. You can then sculpt those branches into a structured screenplay format and export to Fountain for use in professional screenwriting software. For the training data, I spent a month curating 100 films I consider high-quality cinema—Godard, Kurosawa, Brakhage, and others. I built a 1000+ line extraction pipeline using Qwen3-VL to pull narrative structure, characters, and themes from each film with subtitles enabled. From those extractions I generated a 10K example dataset and fine-tuned Qwen2.5-7B-Instruct with a LoRA optimized for probabilistic story branching. The graph visualization is built with React Flow. We've been using it ourselves to break through second-act problems and explore narrative directions we wouldn't have considered otherwise. The branching format forces you to think in possibilities rather than committing too early. You can try it at https://cinegraphs.ai/ — the free tier gives you 3 projects. Would love feedback on the generation quality and whether the graph interface feels intuitive for your workflow.
    • basch16 minutes ago
      Can I ask, did you pull all this from subtitles and scripts?

      Quite frankly, the corpus of film criticism might be a better source. Analysis of context, interpretation, intent, result, success, failure, contention might be more useful to shape a story than the literal story itself. It's asking too much of current gen LLMs to be able to synthesize motif at a higher abstraction. In my experience they get stuck on specific examples and crudely stitching pastiche together, instead of working in the blank empty space between thoughts and ideas.

      I also am unsure that "describe your opening scene" is the best place to start. I may have a story that has a tangible beginning, middle, end, and want to fix certain elements along the way. "This must happen to start act 3" so the story coherently steers correctly towards goals.

    • throwaw1240 minutes ago
      would be nice to see sample conversation shared publicly to understand what it exactly generates. It could be even good for your SEO if you share couple of them