Correction, 2026-10-01: we suggested Kling v3 for cheap variants without saying that on 8frame Kling v3 is image-to-video only and needs a start image. The advice now says so and points text-only variants to Kling 2.6 Pro or Wan 3.0 at 480p.
Meta Muse Video and Veo 3.1 aren't yet competing for the same job, because only one of them can be used. Meta previewed Muse Video on July 7, 2026: native audio, a No. 3 rank in human-preference Elo for text-to-video on Arena at the time, and "coming soon to creators and in Meta AI". As of 2026-09-30 there's no public API, resolution, length or pricing. Veo 3.1 runs today, on Google's API and on 8frame, where 8 seconds at 1080p with audio costs 448 credits. So the practical answer is: produce with what ships, and watch Muse.
Correction, 2026-09-30: the first version of this page said Muse Video clones a likeness from user photos and framed the comparison as avatar video against cinematic video, with a consent controversy on Muse's side; Meta's announcement doesn't describe face-cloning for Muse Video (the episode we can source concerns Muse Image). It also listed Veo 3.1 at 4K 60fps (1080p on 8frame) at $2.24 to $4.48 and Kling at $0.57 to $0.86 (224 to 448 and 57 to 86 credits). This version uses Meta's and Google's pages and 8frame's registry.
TL;DR
- Muse Video: previewed 2026-07-07, native audio, Arena No. 3 for text-to-video at announcement, gaps named by Meta in audio-video sync and fast motion. Still "coming soon"; nothing to test or price.
- Veo 3.1: available now. On 8frame, 4, 6 or 8 seconds at 1080p, 448 credits for 8 seconds with audio, Fast 168, Lite 54. Google lists 1080p and 4K on its own tools and API.
- Both generate audio natively, by their makers' descriptions.
- Choose by availability today, and evaluate Muse on your own briefs once it ships.
What each model is
Meta Muse Video
From Meta's announcement: built on the same pretraining base as Muse Image, "exceptional visual fidelity with native audio support", No. 3 in human-preference Elo for text-to-video on Arena at the time of writing, with current gaps in audio-video synchronization and physically accurate fast motion, and "coming soon to creators and in Meta AI". More in what is Meta Muse Video.
Veo 3.1
Google DeepMind's current Veo release. Google describes native audio (sound effects, ambient noise, dialogue), 1080p and 4K output, 8-second clips, reference images, first and last frame, and scene extension. On 8frame it runs at 1080p (Lite 720p) in 16:9 or 9:16, 4, 6 or 8 seconds from text. More in what is Veo 3.
Side by side
| Meta Muse Video | Veo 3.1 | |
|---|---|---|
| Status (2026-09-30) | Coming soon, preview only | Available |
| Audio | Native, per Meta | Native, per Google |
| Resolution | Not published | 1080p on 8frame; 1080p and 4K on Google's tools |
| Length | Not published | 4, 6 or 8 s on 8frame |
| Price | Not published | 448 credits for 8 s with audio on 8frame; $0.40 a second on the Gemini API |
| API | None published | Gemini API; 8frame canvas and MCP |
| Arena (at Meta's announcement) | No. 3, text-to-video |
A note on likeness
Meta's Muse Image, launched the same day, works from your own photos, and on July 10 Meta removed a feature that let people @-mention public Instagram accounts for image generation, saying it "missed the mark". Whatever model you use, putting a real person's face or voice into generated video needs that person's documented consent.
How to choose
You need to produce this month: Veo 3.1, or another model you can run today. On 8frame: Veo 3.1 for sound in one pass, Seedance 2.5 for takes up to 30 seconds (157 credits for 5 seconds at 720p). For cheap variants of a still you already have, Kling v3 Standard (57 for 5 silent seconds); on 8frame it is image-to-video only, so it needs a start image. For cheap variants from a text prompt alone, Kling 2.6 Pro (47 for 5 silent seconds) or Wan 3.0 at 480p (24 for 5 seconds).
You make content inside Meta's apps and can wait: keep Muse Video on a watch list. When it ships, run the same brief through it and your current model side by side and route future work to whichever delivered.
You need a specific person speaking: on 8frame, Hailuo H3 Max Lip Sync animates a still to an audio track (54 credits for 5 seconds at 768p), with that person's consent.
FAQ
Is Meta Muse Video better than Veo 3.1?
There's no way to test that yet. Meta reported an Arena No. 3 for text-to-video at announcement; Muse Video isn't publicly available, so you can't run your own briefs through it.
Can I use Meta Muse Video on 8frame?
No. It isn't on 8frame and, as of 2026-09-30, not publicly available anywhere. Veo 3.1 is on 8frame today.
Does Muse Video do face-cloning?
Meta's announcement doesn't describe it. Muse Image works from your photos, and Meta pulled a feature that let people @-mention public Instagram accounts for image generation.
Sources
- Meta AI, Introducing Muse Image and Muse Video (2026-07-07): ai.meta.com/blog/introducing-muse-image-muse-video-msl, checked 2026-09-30
- Meta Newsroom, Introducing Muse Image (updated 2026-07-10): about.fb.com/news/2026/07/introducing-muse-image-meta-ai, checked 2026-09-30
- Google DeepMind, Veo: deepmind.google/models/veo, checked 2026-09-30
- Gemini API pricing for Veo 3.1: ai.google.dev/gemini-api/docs/pricing, fetched 2026-09-30
- 8frame prices: the 8frame tool registry, 2026-09-30; Kling v3 image-to-video only, Kling 2.6 Pro and Wan 3.0 prices (calculateKlingV3Cost, calculateKling26Cost, calculateWan3Cost): read 2026-10-01
Produce with what ships, test what's coming on your own briefs.
Try it without signing up: the 8frame canvas opens on a real board. The canvas is free and unlimited; generation is paid from $19/month.