Model Bakeoffrendered on fal · 2026-09-03

InputsOne face, one line › Veo 3.1

Veo 3.1 on One face, one line

Veo 3.1, reference-to-video

What went in

the establishing frame

the establishing frame — no additional references for this input

The woman leans forward slightly, looks into the camera and says: "I told you it would work. I just didn't say when." Then she picks up the mug and takes a sip. The camera holds steady. Natural morning light, room tone, no music.
  • “I told you it would work. I just didn't say when.”
5 s asked9:1686b7da01c0e8

What came out

cost$3.20
wait35s
delivered8.00s
frame720×1280
attempts2 (1 failed)
verdict yellow

Every check on this take

Duration 8.00sexpected 5s (+3.00s)
Audio track 48000 Hz
Loudness -19.4 LUFStrue peak -4.3 dBTP, LRA 9.3 LU
Script lines 1/1all lines heard
Motion 2.15by shot 2.15
Scene changes 0
Attempts 21 failed before success
Wall clock 35s
what whisper heard

I told you it would work. I just didn't say when.

The exact request

Verbatim, as sent to fal-ai/veo3.1/reference-to-video · request 01a0650a-f0e4-7f61-8d45-1fa90c371719. Reference URLs expire; the parameters do not. Send this and you should get something like the video above — that is the point of publishing it.

show the payload
{
  "prompt": "The woman leans forward slightly, looks into the camera and says: \"I told you it would work. I just didn't say when.\" Then she picks up the mug and takes a sip. The camera holds steady. Natural morning light, room tone, no music.",
  "image_urls": [
    "https://v3b.fal.media/files/b/0aa8e1ff/-8_Pz_0Dw2t3Sncaewz-C_face-monologue.jpg"
  ],
  "duration": "8s",
  "aspect_ratio": "9:16",
  "resolution": "720p",
  "generate_audio": true
}

The same input, other models

H3 MaxH3 ref2vKling v3 ProWan 2.7
refused
Seedance 2.0
LTX-2

See all of them side by side →

Everything Veo 3.1 rendered →