Back to News
Advertisement
ddawndrain about 16 hours ago 4 commentsRead Article on github.com

ES version is available. Content is displayed in original English for accuracy.

I've been working on this Claude Code pipeline for making 10-minute movies, gluing together Seedance (videogen), Nano Banana (image gen), and ElevenLabs (voice gen), and using Claude as the director. My repo contributes a markdown playbook and a worked example which should make it easier for other people to make their own short movies. A first-pass movie takes around 2.5 hours of wall clock time and costs around $200, and iterating is easy thanks to intermediate artifacts like storyboards and voice samples.
Advertisement

⚡ Community Insights

Discussion Sentiment

100% Positive

Analyzed from 114 words in the discussion.

Trending Topics

#costs#llm#here#more#https#epoch#model#appreciate#pipeline#interesting

Discussion (4 Comments)Read Original on HackerNews

retelabout 4 hours ago
appreciate the pipeline. it's interesting to read the steps, see the outcome, and know about the costs/time.
dawndrainabout 16 hours ago
In addition to the repo I have a long blog post reflecting on my experience here, including war stories, more detailed costs, and bloopers https://dawndrain.notion.site/Making-AI-Movies-39cbb81dad798...
gulugawaabout 11 hours ago
Its going to cost a lot more than $200 once LLM subsidies run out.
dawndrainabout 5 hours ago
Epoch AI estimates that LLM costs are falling by over 10x per year at fixed intelligence https://epoch.ai/data-insights/llm-inference-price-trends, so I expect prices to fall at this level of quality. I also think there's a good division of labor here where you can use the smartest model to write the script and prompts, and then use a cheaper model for all the glue work and minor revisions that trigger the tool calls.