@sentientqubit I like it! Microslop: Safe slop for creativity, productivity and visual explanations. This is our vision.
English
Nando de Freitas
12.1K posts
Nando de Freitas
@NandoDF
MAI Superintelligence Team. Past: AlphaGo tuning, AlphaCode, Gato, ReST, learning to learn, awareness models, recurrent Gemma, Imagen3, Veo, Genie, MAI models
London, England 가입일 Nisan 2009
822 팔로잉108.9K 팔로워
We may err on not generating the image you need, an realise there is room for improvement, but safety is our topmost priority
English
Nando de Freitas 리트윗함
Microsoft’s MAI-Image-2 enters the top three AI image generators in the world thenextweb.com/news/microsoft… via @thenextweb
English
and they can also mix concepts compositionally
“Un elefante rosa en hermoso Bariloche, Argentina”
English
Mai-image: “Un pajaro Jabirú en Argentina”
Amazing how this and many models have so many visual concepts in them.
English
The grok answer addresses your question about public licenses. I'd like to emphasise that the photographs that we purchase are essential to generate high quality artistic images. Now more than ever, photographers can profit by selling their images to companies like ours or many startups.
English
@NandoDF Thank you! What does ‘legally available data’ mean?
English
playground.microsoft.ai is now available in the USA and soon all over the world.
Today we unveil our image generation models. Soon everyone will have access to many more models to play with in our playground. Enjoy!
Wanna help build these models? JoinAITeam@microsoft.com - In particular, I'm looking for exceptional data, research and infra engineers.
English
@ednewtonrex @grok please explain the different image licenses that exist for images on the web, and how to comply with them in order to train an image generation model.
English
Nando de Freitas 리트윗함
Microsoft just shipped MAI-Image-2.
They built it by talking to working photographers and designers first. Not researchers.
Enhanced photorealism that passes for real photos. In-image text that doesn't break. Complex scenes that hold together.
Available now in MAI Playground.
Rolling out on Copilot and Bing Image Creator.
API access opening to developers on Microsoft Foundry soon.
Satya Nadella@satyanadella
Great to see our new image model from our Superintelligence team rolling out in Copilot and coming soon to Foundry for enterprise customers.
English
@ednewtonrex Legally available data and licensed data. For us trust is important.
English
@Zero2Ten I agree. We preferred to err on the side of safety, but are working on it. Thanks for trying it and your great feedback 🙏
English
@NandoDF The model is great, but the prompt moderation/censorship is atrocious.
English
Nando de Freitas 리트윗함
Great to see our new image model from our Superintelligence team rolling out in Copilot and coming soon to Foundry for enterprise customers.
Mustafa Suleyman@mustafasuleyman
Our new image generator MAI-Image-2 is out! Available now on MAI Playground for everything from lifelike realism to detailed infographics. Our team has been pushing immensely hard for this release, and we are now among the top models out there: #3 family on @arena. Check out the details in our blog: microsoft.ai/news/introduci… It's shipping soon in Copilot and Bing Image Creator, as well as Microsoft Foundry. Really proud of our progress on models and products - stay tuned for new releases and come join us on our Superintelligence mission!
English
@victor_UWer Data work, and people with no ego, who know that working in teams is more effective ... and more fun.
English
Super proud of our small image-gen team that in less than 6 months achieved this amazing result.
.... and they're already cooking the next one, but I'm encouraging them all to take a vacation😎
Microsoft AI@MicrosoftAI
Meet MAI‑Image‑2. Built with creatives, for real creative work. Ranked #5 on @arena’s text‑to‑image leaderboard. Available now: msft.it/6014QUCBe
English
Nando de Freitas 리트윗함
Let’s dive deeper into the massive improvements between MAI-Image-2 vs. MAI-Image-1 by @MicrosoftAI.
MAI-Image-2 shows significant gains across all sub-categories for Text-to-Image:
Gains across all 7 sub-categories in order of magnitude:
- Text Rendering (+115 pts)
- Portraits (+105 pts)
- Product, Branding & Commercial Design (+102 pts)
- Photorealistic & Cinematic Imagery (+97 pts)
- 3D Imaging & Modeling (+92 pts)
- Art (+87 pts)
- Cartoon, Anime & Fantasy (+81 pts)
Arena.ai@arena
MAI-Image-2 debuts at #5 in the Image Arena! Highlights: - #5 in Text-to-Image overall - #5 for 3D Imaging & Modeling, Cartoon, Anime & Fantasy, Photorealistic & Cinematic Imagery, Art and Portraits - #6 for Product, Branding & Commercial Design Congrats to the @MicrosoftAI team on this milestone!
English