Same creator, two video-model shootouts six weeks apart, opposite verdicts on Gemini Omni Flash: test conditions (resolution cap, start-frame use, prompt specificity) decide rankings more than the models

Creator-reported: statements attributed to the creator(s) come from discovery's reading of auto-caption transcript excerpts; the curator checked the video metadata (title, date, length) and any cited documentation pages, not the spoken claims. Nothing was tested; promotional or sponsored framing applies where stated. Two Higgsfield-hosted comparisons by Youri van Hofwegen. Aug 2026 (R2): five models at each model's MAX settings (Seedance 2.0 native 4K 15 s, Sora 2 Pro Max 12 s 1080p, Kling 3.0 4K, Happy Horse 1080p, Gemini) on a talking-head lip-sync test and a breakdance motion test, scored 1-10. Verdict: Seedance top on both; "Gemini is the real surprise" top-two while cheapest (30 credits vs Kling 90, Happy Horse 68, Sora 153, Seedance 330 - Seedance partly because of native 4K); Kling 3 lip sync "pretty bad" (5/10); Sora bottom. Sep 2026 (R1): four models BLIND-labelled A-D, all capped at 10 s, 16:9, 720p with audio "otherwise this becomes a test of which model has the biggest resolution", five prompt rounds (multi-cut dialogue with word-for-word lines, cel-shaded anime action, physics/card castle, instruction-specific candles, image-to-video from a shared start frame). Ranking: Seedance 2.5 first (only model to follow the prompt in all five rounds; 65 credits), Kling 3 second (17.5 credits; best emotional line delivery, good start-frame hold; weak on stylised/physics), Happy Horse third (25 credits; consistent, good anime/lighting; failed start-frame turn), Gemini Omni Flash last (30 credits; good physics and lip sync but ignored specific instructions; no start-frame support so it "fails by default"). Takeaways for running your own test: hold resolution/duration/audio constant, generate before watching and label blind, write dialogue verbatim so lip sync can be checked, separate text-to-video from reference/start-frame tests, and report credits per run alongside quality - the same model's value flips with the cap. Curator qualifications: Creator-reported: statements attributed to the creator(s) come from discovery's reading of auto-caption transcript excerpts; the curator checked the video metadata (title, date, length) and any cited documentation pages, not the spoken claims. Nothing was tested; promotional or sponsored framing applies where stated. All cited documentation excerpts located on the live pages by curator GET; video metadata checked where cited. Creator video (auto-caption transcript excerpts read by discovery); not curator-verified beyond video metadata. Limitations: Creator statements were read by discovery from auto-captions (ASR); not re-verified by the curator. Both videos are Higgsfield affiliate content (R1 description flags sponsorship; both link youricreates.com/Higgsfield and pitch a bonus package), single generation per model per prompt (no seeds/repeats), creator-judged scores; prompts not published outside the paid bonus pack; R2 compares Seedance 2.0 while R1 uses 2.5, and R2 does not name the Gemini model version in the excerpted text. Credits are Higgsfield platform credits, not vendor API prices. Auto-caption ASR spells models phonetically (Cedance, Cling, Higsfield). Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.

{
  "citations": [
    {
      "url": "https://www.youtube.com/watch?v=v6bqRIjwddE",
      "note": "1:40; 0:30; 16:13; 12:12; 17:15"
    },
    {
      "url": "https://www.youtube.com/watch?v=PsJLCiWnZhM",
      "note": "1:40; 0:30; 16:13; 12:12; 17:15"
    }
  ]
}