AINewsnow

Runway Details Serving 57B-Parameter Character Model in Real Time on Half the GPUs

This story is from 2026-11-05. It is preserved in the archive; the latest stories are on the live feed.

Runway Research described how it serves its 57-billion-parameter model behind Runway Characters fast enough for live conversation, using half the GPUs it launched with.

Read the full story at Runway Research ↗

Timeline · 1 report

  1. 2026-11-05 16:43 · Runway Research
    Real-Time Video at Batch Size One

More stories

  1. Generate longer AI videos with Omni 1.2 Flash in Google Vids at no cost. — Google: The Keyword
  2. We Used to Have Time to Be Wrong — DEV Community — AI
  3. The World Model Trilemma: Pixels, Splats, and Symbolic Physics — DEV Community — Machine Learning
  4. Why your AI video prompts produce morphing hands and flickering faces (and the syntax that fixes it) — DEV Community — Machine Learning
  5. Kling 4.0 Omni Reference: the real limits, the prompt structure, and the photos behind each clip — DEV Community — Machine Learning
  6. Krea Video brings Veo 3.1, Sora 2, Seedance and Kling into one browser workspace — AlphaSignal
  7. Meta's Muse Generates Real-Time Talking Avatars From Any Portrait — AlphaSignal
  8. Election backlash against data centers rattles AI stocks and expansion plans — TechCrunch AI

Get the daily brief of stories like this at 6:30 every morning →