New Meta image and video AI tools let you turn Instagram into your creative status board

Published:

Meta has been cooking something, and today, it finally put it on the table. On July 7, 2026, Meta Superintelligence Labs launched Muse Image and Muse Video (in preview), its first indoor media production models.

The release comes with a few features that are really hard to argue with.

Introducing Muse photo and Muse video, the first media production models developed by Meta Superintelligence Labs.

Muse Image is our most advanced image creation model yet. Follows directions faithfully, organizes accurately, integrates multiple directions, and draws… pic.twitter.com/byNpQZO1RW

— AI at Meta (@AIatMeta) July 7, 2026

What makes Muse Image truly unique?

You can mention a public Instagram account (using ‘@’) directly in your notification, and Photo Muse will use images from that profile as visual references for your design.

So, if you are looking for a custom event invitation that showcases the beauty of your brand, or you are creating a creative concept with an editor, you can simply mark the required account.

It is a truly novel feature, integrating image generation directly into the social graph Meta has dominated for two decades.

Besides, Muse Image works more like an agent than a traditional text-to-image model.

Exciting news: Meta’s Muse image just claimed #2 in Image Arena!

Muse Image from @AIatMeta is now ranked second behind OpenAI’s GPT Image 2, outperforming Nano Banana, Grok Imagine, MAI Image, and many other leading image models.

Ranked #2 across the board: Text-to-Image,… pic.twitter.com/577214iiqj

— Arena.ai (@arena) July 7, 2026

How exactly does it stack up against the competition?

Rather than generating a single output based on your input, it thinks about the request, searches the web for the correct context, writes and executes code if accuracy is important (for charts or QR codes), and adjusts the output itself before giving you the result.

On Arena.ai’s people’s favorite leaderboard, which ranks models based on real user votes, Muse Image holds second place among text-to-image, single-image editing, and multi-image editing as of July 5, 2026.

Only OpenAI’s GPT Image 2 ranks higher, at 1,385, compared to Muse Image’s 1,280. Meanwhile, Google’s Nano Banana 2 (1,270) and MAI Image 2.5 (1,257) trail behind.

Muse Image is free for daily use on the Meta AI app (meta.ai), Instagram Stories (US), and WhatsApp in select countries. Facebook support is coming soon. For those who need high volume creations, the subscription phase enables that.

1/ today, with the presentation of the museum image and video preview of the museum, we wanted to share more samples from the model and the details of the research and the eval results

see some samples from the museum video below

(and see the research blog series!) pic.twitter.com/7yVUOolkG4

— Alexandr Wang (@alexandr_wang) July 7, 2026

What about the Muse video?

Built on the same training foundation as Muse Image, Muse Video supports native audio, producing audio alongside visuals.

On Arena.ai, it is already ranked #3 for text-to-video popularity. Meta has been upfront that the audio and video sync and motion physics still need work, which is why this is a preview and not a full launch.

Muse Video is coming soon to creators and Meta AI, with no firm date attached yet.

For me, Muse Image and Muse Video are not just product releases. They represent Meta cutting ties with third-party AI vendors it has relied on for years. Building in-house provides fast Meta iteration, tight integration, and control over the tools marketers increasingly monetize.

Related articles

spot_img

Recent articles

spot_img