Matt Deitke's banner

Matt Deitke

@mattdeitke • 15,758 subscribers

AI Researcher in TBD Lab @ Meta Prev: Co-founder @ Vercept (acq. Anthropic), researcher @allen_ai, PhD dropout @uwcse

Shorts

Muse on Mac has instant voice dictation built right in! 🎧 Just hold fn and start talking. No separate dictation app, no setup. It’s insanely fast. It's so much faster than typing and nearly instant. It's such a nice use of the multimodal capabilities of the models.

Muse on Mac has instant voice dictation built right in! 🎧 Just hold fn and start talking. No separate dictation app, no setup. It’s insanely fast. It's so much faster than typing and nearly instant. It's such a nice use of the multimodal capabilities of the models.

207,967 просмотров

Introducing Objaverse, a massive open dataset of text-paired 3D objects! Nearly 1 million annotated 3D objects to pave the way to build incredible large-scale 3D generative models: 🧵👇 🤗 Hugging Face: 📝ArXiv: #CVPR2023

Introducing Objaverse, a massive open dataset of text-paired 3D objects! Nearly 1 million annotated 3D objects to pave the way to build incredible large-scale 3D generative models: 🧵👇 🤗 Hugging Face: 📝ArXiv: #CVPR2023

1,601,334 просмотров

Computer use has gotten incredibly good in the past year, and it works great in Muse out of the box! Muse ships with a browser that it uses for tasks that it can't solve with an API or connector. At this point, it's hard to come up with tasks that computer use models aren't able to solve well, even when spanning into a 1hr+ time horizon (see the progress on OSWorld 2.0!) Here's a sped up version of it working for finding the cheapest flights from SF -> NYC!

Computer use has gotten incredibly good in the past year, and it works great in Muse out of the box! Muse ships with a browser that it uses for tasks that it can't solve with an API or connector. At this point, it's hard to come up with tasks that computer use models aren't able to solve well, even when spanning into a 1hr+ time horizon (see the progress on OSWorld 2.0!) Here's a sped up version of it working for finding the cheapest flights from SF -> NYC!

36,895 просмотров

Today, we're introducing Muse Image and Muse Video! 🖌️ Muse Image is an agentic image generation model with strong integrations with coding, search, and self-refinement! Muse Video is an early preview of our video generation efforts with strong audio and video generation abilities! Details below! 🧵👇

Today, we're introducing Muse Image and Muse Video! 🖌️ Muse Image is an agentic image generation model with strong integrations with coding, search, and self-refinement! Muse Video is an early preview of our video generation efforts with strong audio and video generation abilities! Details below! 🧵👇

53,411 просмотров

The multimodal grounding abilities combined with web development in Muse Spark are so good! 🎨 Prompt: <image of dogs> Make a website where you point to each of the dogs and say their breed.

The multimodal grounding abilities combined with web development in Muse Spark are so good! 🎨 Prompt: <image of dogs> Make a website where you point to each of the dogs and say their breed.

17,985 просмотров

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Muse Wonder can generate lots of fun 3D games in the web! Even generating the Jolly character 🥳 😁

Muse Wonder can generate lots of fun 3D games in the web! Even generating the Jolly character 🥳 😁

31,011 просмотров • 7 дней назад

Muse has a huge advantage with Instagram. Asked for the best hikes in San Francisco and got a Reel for every recommendation, playable right in the answer. The amount of real-world knowledge sitting inside Instagram is insane for most consumer use cases. 🐣

Muse has a huge advantage with Instagram. Asked for the best hikes in San Francisco and got a Reel for every recommendation, playable right in the answer. The amount of real-world knowledge sitting inside Instagram is insane for most consumer use cases. 🐣

119,452 просмотров • 1 месяц назад

Real-time conversations with Muse is one of the most mind blowing experiences I've tried this year. It felt unreal. A lot of the voice and visual interactions were pretty spot on to mimic how people naturally talk. The tech behind this work is so impressive, especially to make it realtime and cheap enough to serve at scale. Excited for you all to try it!

Real-time conversations with Muse is one of the most mind blowing experiences I've tried this year. It felt unreal. A lot of the voice and visual interactions were pretty spot on to mimic how people naturally talk. The tech behind this work is so impressive, especially to make it realtime and cheap enough to serve at scale. Excited for you all to try it!

50,271 просмотров • 26 дней назад

Muse on Mac can now take over your computer and do things for you! Local computer use has shipped at Connect! 🥳 This is what I worked on at my startup 2 years ago with many great folks before joining Meta. Very excited to see it now shipped for the masses!

Muse on Mac can now take over your computer and do things for you! Local computer use has shipped at Connect! 🥳 This is what I worked on at my startup 2 years ago with many great folks before joining Meta. Very excited to see it now shipped for the masses!

48,924 просмотров • 26 дней назад

Ray-Ban Meta Glasses are so great for events and adventures outside. This commercial is so good and captures it well. And with the multimodal models and new Muse integration, it'll be incredible! 😎

Ray-Ban Meta Glasses are so great for events and adventures outside. This commercial is so good and captures it well. And with the multimodal models and new Muse integration, it'll be incredible! 😎

15,123 просмотров • 11 дней назад

Turning any book into a chapter-by-chapter podcast series in Muse might be my favorite new use case. 🔉 It brings together so many things: - Long context agents that can reason through very long books - Subagents that go chapter by chapter to operate in parallel - Podcast generation - Website generation to build an artifact website that persists so I can easily go back to it Just upload the book and ask it to generate podcasts of every chapter of the book and then build an artifact out of it. It did this all from one prompt. No new plugins or extensions required. It’s so mind blowing to me that this works so well! Below is a link to the generated site 👇

Turning any book into a chapter-by-chapter podcast series in Muse might be my favorite new use case. 🔉 It brings together so many things: - Long context agents that can reason through very long books - Subagents that go chapter by chapter to operate in parallel - Podcast generation - Website generation to build an artifact website that persists so I can easily go back to it Just upload the book and ask it to generate podcasts of every chapter of the book and then build an artifact out of it. It did this all from one prompt. No new plugins or extensions required. It’s so mind blowing to me that this works so well! Below is a link to the generated site 👇

36,424 просмотров • 1 месяц назад

Introducing Objaverse-XL, an open dataset of over 10 million 3D objects! With it, we train Zero123-XL, a foundation model for 3D, observing incredible 3D generalization abilities: 🧵👇 📝 Paper:

Introducing Objaverse-XL, an open dataset of over 10 million 3D objects! With it, we train Zero123-XL, a foundation model for 3D, observing incredible 3D generalization abilities: 🧵👇 📝 Paper:

368,893 просмотров • 3 лет назад

Excited to share Muse Spark! It's a strong natively multimodal model with many surprising properties that emerged. Here, the model is able to use Python tools to make a playable Sudoku game on the web from an image input of the board. ✨

Excited to share Muse Spark! It's a strong natively multimodal model with many surprising properties that emerged. Here, the model is able to use Python tools to make a playable Sudoku game on the web from an image input of the board. ✨

29,722 просмотров • 6 месяцев назад

Больше нет контента для загрузки