VQV Signal

CREATOR TOOLS WATCH TECHNICAL

HeyGen's Decision 1 video-understanding model now available on OpenRouter and AI/ML API

According to a post on X by @ModelsOnOR, OpenRouter said HeyGen's Decision 1 model answers questions about video, audio, images and text in a single request. It returns a calibrated probability for each answer in about 100 ms, with up to 400 questions per call. Launch pricing is 50% off through Dec 1, at $0.07 per 1M input tokens, or $0.042 for the Turbo version, with output listed as free. AI/ML API also demoed the model turning gestures and emotions in a video into emojis.

Source: Scobleizer News Pulse (X) · x.com Found via Robert Scoble’s Scobleizer News Pulse · a post on X, not independently verified Published 2026-11-20T23:45:00+00:00 Detected 2026-11-21T09:05:12+00:00
View original source

According to a post on X by @ModelsOnOR, OpenRouter said HeyGen's Decision 1 model answers questions about video, audio, images and text in a single request. It returns a calibrated probability for each answer in about 100 ms, with up to 400 questions per call. Launch pricing is 50% off through Dec 1, at $0.07 per 1M input tokens, or $0.042 for the Turbo version, with output listed as free. AI/ML API also demoed the model turning gestures and emotions in a video into emojis.

AI-assisted summary based on the listed source.

OpenRouter said HeyGen's Decision 1 answers questions about video, audio, images and text in one request, returning a calibrated probability for each answer in about 100 ms, with up to 400 questions per call. Launch pricing is 50% off through Dec 1 ($0.07 per 1M input tokens, or $0.042 for Turbo; output is free)....

Developers and businesses working with video, audio or images could test one model that handles several input types in a single request, with promotional pricing running through Dec 1. The per-answer probability output may matter to teams that need confidence scores alongside answers. Watch whether pricing holds after the promotion ends and how the model performs on real workloads.

Signal Strength 88% Technical label WATCH Public Interest 54 Category CREATOR TOOLS Reader Depth TECHNICAL

Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.

Public Interest components
Recognizable Entity Score 0 Practical Impact Score 74 Novelty Interest Score 94 Consequence Score 62 Curiosity Score 52 Shareability Score 58

VQV surfaced this signal because it is recent, relevant to AI Chips, connected to Scobleizer News Pulse (X).