Routing 1000+ video models: filter with code first, let the LLM pick from a shortlist
Vaciuum1 · reddit · 2026-10-10
A backend engineer on a commercial AI video agent shares how they handle model routing. With 1000+ video models, letting the LLM pick freely means it confidently grabs models that can't even accept the user's reference image or 9:16 aspect ratio.
Their approach is two-layered:
- Code-first filtering: before the LLM sees anything, every model that can't handle the inputs (reference images, length, aspect ratio, audio) is removed;
- Scoring + shortlist: remaining models are scored by release date, description, and user memory, and the LLM only picks from a shortlist;
- User confirmation: a draft with the chosen model and cost is shown before anything runs.
The author asks the community: rules + scoring like us, or something learned?
Related event: Dev Shares Routing Tips for 1000+ AI Video Models(2 posts)→
More from coding & agent
- Gemini 4 Argon tops deepswe at 77.9% and automationbench, still locked to trusted testers — weswinder · 2026-10-10
- TWIML podcast: TypeSafe's Jev model bets on machine-native intelligence over LLMs — samcharrington · 2026-10-10
- mitsuhiko: template engines like jinja2 are unsafe for untrusted input without OS-level isolation — mitsuhiko · 2026-10-10
- Decision models with non-deterministic rules could radically improve AI codegen — rickasaurus · 2026-10-10
- What if the Cloudflare dashboard was an infinite canvas? A demo — round · 2026-10-10
- Anthropic AI model submitted fabricated homicide tip to Philadelphia police during testing — rohanpaul_ai · 2026-10-10