正在加载视频...

视频加载失败

返回首页

We built high-throughput materials labs in Menlo Park to create a loop between experiments and models. The labs generate fresh data, the models learn from it, and then help us decide what to try next. Using only 1,300 H200s, plus months of our experimental data, we mid-trained and RL’d...

LiamFedus's profile picture

Liam Fedus

39,647 subscribers

1,058,825 次观看 • 15 小时前 •via X (Twitter)

新闻政治教育科学技术
LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

I announced a Critical Raw Materials Act in my 2022 State of the Union address. And this year, we made it happen. It is crucial to our security of supply. And it helps us and our reliable partners to strengthen our industries and create good jobs. Join me on 13/09 for #SOTEU
0:16

I announced a Critical Raw Materials Act in my 2022 State of the Union address. And this year, we made it happen. It is crucial to our security of supply. And it helps us and our reliable partners to strengthen our industries and create good jobs. Join me on 13/09 for #SOTEU

Ursula von der Leyen

113,673 次观看 • 3 年前

Today we're dropping the "beta" tag from Adaptyv, launching our new website and announcing our $8M seed round. When we started Adaptyv a few years ago, our core belief was: AI models for biology are only as good as the experimental data they're trained on and the hypotheses they can test in the real world. Now, after a year of working with many great partners, we’ve scaled our infrastructure to the point that we're now open to anyone who wants to use our platform! Overall, this year, over 30 companies started using Adaptyv to validate their protein designs - from some of the biggest pharmas to frontier AI labs to many, many techbio startups. We've run hundreds of experiments, tested well over 10,000 proteins this year and are generating the data that validates the best AI models currently in development.
0:11

Today we're dropping the "beta" tag from Adaptyv, launching our new website and announcing our $8M seed round. When we started Adaptyv a few years ago, our core belief was: AI models for biology are only as good as the experimental data they're trained on and the hypotheses they can test in the real world. Now, after a year of working with many great partners, we’ve scaled our infrastructure to the point that we're now open to anyone who wants to use our platform! Overall, this year, over 30 companies started using Adaptyv to validate their protein designs - from some of the biggest pharmas to frontier AI labs to many, many techbio startups. We've run hundreds of experiments, tested well over 10,000 proteins this year and are generating the data that validates the best AI models currently in development.

Adaptyv Bio

11,484 次观看 • 1 年前

We recently discovered the ultimate workflow to restore old footage to something close to digital high-quality footage. The method works better than we thought, and it utilizes OpenAI GPT-Image 2 + seedance2.0 2.0 + Topaz Labs Astra…. See the step-by-step workflow in 🧵 Shoutout to Adam Nieri from our team for the real footage + test!
0:15

We recently discovered the ultimate workflow to restore old footage to something close to digital high-quality footage. The method works better than we thought, and it utilizes OpenAI GPT-Image 2 + seedance2.0 2.0 + Topaz Labs Astra…. See the step-by-step workflow in 🧵 Shoutout to Adam Nieri from our team for the real footage + test!

Curious Refuge

35,729 次观看 • 3 个月前

In our constant commitment to bring value to holders, we’re excited to announce the addition of image generation to our dApp on April 3rd Next week, as part of our journey towards decentralization, we are starting to integrate our open-source-based AI model into the Solana blockchain. This marks the beginning of our transition from relying on third-party APIs to utilizing fine-tuned open source AI Models integrated with Solana, enhancing our dApp’s security and autonomy.
0:13

In our constant commitment to bring value to holders, we’re excited to announce the addition of image generation to our dApp on April 3rd Next week, as part of our journey towards decentralization, we are starting to integrate our open-source-based AI model into the Solana blockchain. This marks the beginning of our transition from relying on third-party APIs to utilizing fine-tuned open source AI Models integrated with Solana, enhancing our dApp’s security and autonomy.

PrimeCircle Studio

17,451 次观看 • 2 年前

Today, we’re sharing the history of Foundation Labs, written from our perspective as founders. What we thought, what we set out to build, and what we learned from six years on the frontier of crypto, culture, commerce, and art. We’re publishing this now while the experience is still fresh. It was important to us that a clear record exists in our own voice, rather than leaving our story to be interpreted by others over time. Our aim is to contribute to the commons. We believe there are many future chapters still to be written with this technology, and we want future founders to be able to reference and build on our experience—to start on second base. We’ll keep this online as a resource for the foreseeable future. To the artists, collectors, investors, and former teammates—thank you for being part of the journey.
0:20

Today, we’re sharing the history of Foundation Labs, written from our perspective as founders. What we thought, what we set out to build, and what we learned from six years on the frontier of crypto, culture, commerce, and art. We’re publishing this now while the experience is still fresh. It was important to us that a clear record exists in our own voice, rather than leaving our story to be interpreted by others over time. Our aim is to contribute to the commons. We believe there are many future chapters still to be written with this technology, and we want future founders to be able to reference and build on our experience—to start on second base. We’ll keep this online as a resource for the foreseeable future. To the artists, collectors, investors, and former teammates—thank you for being part of the journey.

kayvon

31,575 次观看 • 6 个月前

From Nasheed selling off our airport to Indian company GMR, to Maldivians fighting hard to take it back, and winning, at great cost. Since then, through our own loans and the efforts of three governments over nearly a decade, we built it back. Now, we’re ready to open our own terminal. Built by us. Owned by us. A journey worth remembering. 🇲🇻
0:15

From Nasheed selling off our airport to Indian company GMR, to Maldivians fighting hard to take it back, and winning, at great cost. Since then, through our own loans and the efforts of three governments over nearly a decade, we built it back. Now, we’re ready to open our own terminal. Built by us. Owned by us. A journey worth remembering. 🇲🇻

Midhuam Saud(米渡）🇲🇻

79,157 次观看 • 1 年前

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.
0:14

1/ Gemini 2.5 is here, and it’s our most intelligent AI model ever. Our first 2.5 model, Gemini 2.5 Pro Experimental is a state-of-the-art thinking model, leading in a wide range of benchmarks – with impressive improvements in enhanced reasoning and coding and now #1 on Arena by a significant margin. With a model this intelligent, we wanted to get it to people as quickly as possible. Find it on Google AI Studio and in the Google Gemini for Gemini Advanced users now – and in Vertex in the coming weeks. This is the start of a new era of thinking models – and we can’t wait to see where things go from here.

Sundar Pichai

864,698 次观看 • 1 年前

We’re excited to announce our integration with SKALE, a high-performance, zero-gas blockchain purpose-built for speed, scale, and security. This partnership strengthens our infrastructure as we continue building transparent, trust-based systems for decentralized science. We’re excited about what this unlocks for researchers, contributors, and the future of data integrity in DeSci. 👀 Look out for more on how we’re using SKALE in the AxonDAO ecosystem.
0:15

We’re excited to announce our integration with SKALE, a high-performance, zero-gas blockchain purpose-built for speed, scale, and security. This partnership strengthens our infrastructure as we continue building transparent, trust-based systems for decentralized science. We’re excited about what this unlocks for researchers, contributors, and the future of data integrity in DeSci. 👀 Look out for more on how we’re using SKALE in the AxonDAO ecosystem.

AxonDAO

33,926 次观看 • 1 年前

(1/n) 🚀 With FastVideo, you can now generate a 5-second video in 5 seconds on a single H200 GPU! Introducing FastWan series, a family of fast video generation models trained via a new recipe we term as “sparse distillation”, to speed up video denoising time by 70X! 🖥️ Live demo: (Thanks to @gmicloud for the support!) 🔗 Blog: 🔓 We fully open-source our models, code, and data with Apache-2.0 licenses
0:13

(1/n) 🚀 With FastVideo, you can now generate a 5-second video in 5 seconds on a single H200 GPU! Introducing FastWan series, a family of fast video generation models trained via a new recipe we term as “sparse distillation”, to speed up video denoising time by 70X! 🖥️ Live demo: (Thanks to @gmicloud for the support!) 🔗 Blog: 🔓 We fully open-source our models, code, and data with Apache-2.0 licenses

Hao AI Lab

78,660 次观看 • 1 年前

HOLY MOLY: Aikido got GPT-6 Astra in advance to run it on our Cybersecurity benchmark, it crushed EVERY other model! - At pass@3 it rediscovered 29/32 CVEs, the highest recall we've ever recorded and 4 more than GPT-5.6-Sol - Even at pass@1, it has 75% recall. The model is VERY consistent - The performance however come at a high price (literally). The three runs cost us almost $4,000 Astra is now the #1 model on the benchmark Debarshi and I built, and by a LOT 1/3 🧵
0:12

HOLY MOLY: Aikido got GPT-6 Astra in advance to run it on our Cybersecurity benchmark, it crushed EVERY other model! - At pass@3 it rediscovered 29/32 CVEs, the highest recall we've ever recorded and 4 more than GPT-5.6-Sol - Even at pass@1, it has 75% recall. The model is VERY consistent - The performance however come at a high price (literally). The three runs cost us almost $4,000 Astra is now the #1 model on the benchmark Debarshi and I built, and by a LOT 1/3 🧵

pilvar (Philippe Dourassov)

48,998 次观看 • 11 天前

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷
0:13

A Letter to Our Community: The Road Ahead for Robotics To our Community and Partners, As we step into 2026, our mission at Axis is clearer than ever: Constructing the definitive End-to-End Scaling Layer for Robotics. Our goal is to accelerate the transfer of diverse human intelligence into Robotics General Intelligence (RGI). By owning the critical path of intelligence creation, we are turning the physical limitations of robotics into a scalable, software-driven future. Here is our strategic outlook and roadmap for the year ahead. The Core Thesis: Simulation is the Only Way Out The path to RGI is currently blocked by Data Scarcity, Generalization Fragility, and Hardware Fragmentation. At Axis, we believe Simulation is the only way out. Our Simulation Data Platform and Data Augmentation Engine transform raw data into "Synthetic Gold". Backed by academic milestones like Roboverse, Skill Blending, and GraspVLA, we have proven that pure simulation can achieve the generalization required for the real world. We don’t just collect data; we architect it. The Engine: Why Crypto? We believe RGI should come from all, not a few. Crypto is not just a feature; it is the primitive that powers our entire ecosystem flywheel: - Incentive Mechanism: Democratizing contribution and rewarding the trainers and developers. - Assetization: Turning proprietary data and refined models into liquid, ownable assets. - Verifiable Workflow: We are opening the "Black Box" of AI. By bringing total transparency to the Task Generation → Data Collection → Model Training pipeline, we ensure every byte of intelligence is verifiable, traceable, and secure. 2026 Strategic Deliverables This year, we are committed to delivering three foundational pillars: - The World's Largest Training Dataset for Robots: A robot training set—diverse, high-quality interaction data at an unprecedented scale. - A Robotics Foundation Model: A universal robotic brain trained on our pure simulation and synthetic data, capable of robust cross-embodiment transfer and open-world adaptability. - Evolvable Robot Hardware: Robots deployed with Axis models that autonomously evolve through continuous interaction, turning every deployment into a self-improving node within our RGI network. The Ultimate Vision We are building more than models; we are architecting the Distributed Machine Economy. A future where every dataset, model, and robotic embodiment is a verifiable asset in a global, autonomous network. Thank you for building the future of intelligence with us✌️📷

Axis Robotics

28,096 次观看 • 8 个月前

To replace animal testing with AI, we need MASSIVE human datasets. Today, we're thrilled to share Axiom's new data exploration tool, providing the ability to visually explore the world's largest primary human liver toxicity dataset. Built with Axiom's proprietary wetlab protocols, our dataset includes detailed liver toxicity profiles for over 100,000 distinct molecules. The key to this dataset is our ability to do high-throughput, multiplexed high-content screening with primary human liver cells. Traditionally, toxicity assays either sacrifice throughput or sacrifice biological relevance (using easy-to-grow immortalized cell lines instead of real human cells). We managed to combine throughput, physiological relevance, and multiplexing in one platform. The assays run in a high throughput format using automation, meaning thousands of compound-dose conditions can be tested in one experiment. We achieved this using pooled primary human hepatocytes, which are often fragile and expensive. By systemizing our automation and quality control processes, we were able to run over 120+ batches on the same donor pool with incredible reproducibility and consistency. We did this while integrating many readouts per well, whereas many existing toxicity assays only do a single readout. Our multiplexed approach provides far more data per experiment enabling us to measure 10-20 different toxicity phenotypes such as apoptosis, necrosis, mitochondrial fission, endoplasmic reticulum stress, stress granule formation, microtubules, and more all from a single well on a 384-well plate! The combination of scale, high content information, and data quality is exactly what is needed to train highly accurate AI models in biology. If you're interested, please explore the dataset in the comments below and let me know if you want to chat about the details!
0:19

To replace animal testing with AI, we need MASSIVE human datasets. Today, we're thrilled to share Axiom's new data exploration tool, providing the ability to visually explore the world's largest primary human liver toxicity dataset. Built with Axiom's proprietary wetlab protocols, our dataset includes detailed liver toxicity profiles for over 100,000 distinct molecules. The key to this dataset is our ability to do high-throughput, multiplexed high-content screening with primary human liver cells. Traditionally, toxicity assays either sacrifice throughput or sacrifice biological relevance (using easy-to-grow immortalized cell lines instead of real human cells). We managed to combine throughput, physiological relevance, and multiplexing in one platform. The assays run in a high throughput format using automation, meaning thousands of compound-dose conditions can be tested in one experiment. We achieved this using pooled primary human hepatocytes, which are often fragile and expensive. By systemizing our automation and quality control processes, we were able to run over 120+ batches on the same donor pool with incredible reproducibility and consistency. We did this while integrating many readouts per well, whereas many existing toxicity assays only do a single readout. Our multiplexed approach provides far more data per experiment enabling us to measure 10-20 different toxicity phenotypes such as apoptosis, necrosis, mitochondrial fission, endoplasmic reticulum stress, stress granule formation, microtubules, and more all from a single well on a 384-well plate! The combination of scale, high content information, and data quality is exactly what is needed to train highly accurate AI models in biology. If you're interested, please explore the dataset in the comments below and let me know if you want to chat about the details!

Brandon White

25,117 次观看 • 1 年前

Stormwind was the first city we ever built. Our lead programmer had it load and play the song "We built this city" to our surprise when we first loaded in. The first version of Stormwind had no districts or canals. The framerate was below terrible. Then I thought of Venice. I grabbed a map of the city and slapped it down on the table the next morning. THIS, I said, is how we'll partition the city, drawing only the district you can see, separated by canals. It worked, and Stormwind was born.
0:15

Stormwind was the first city we ever built. Our lead programmer had it load and play the song "We built this city" to our surprise when we first loaded in. The first version of Stormwind had no districts or canals. The framerate was below terrible. Then I thought of Venice. I grabbed a map of the city and slapped it down on the table the next morning. THIS, I said, is how we'll partition the city, drawing only the district you can see, separated by canals. It worked, and Stormwind was born.

Grummz

245,446 次观看 • 6 个月前

We get portrayed as vigilantes or racists or far right nazis we have heard It all but nothing will stop us from trying to make sure our woman and children are safe on our streets we will try and make sure everyone is aware of the real dangers our city is in 💯🙏🏻🏴󠁧󠁢󠁳󠁣󠁴󠁿
0:16

We get portrayed as vigilantes or racists or far right nazis we have heard It all but nothing will stop us from trying to make sure our woman and children are safe on our streets we will try and make sure everyone is aware of the real dangers our city is in 💯🙏🏻🏴󠁧󠁢󠁳󠁣󠁴󠁿

P Maguire

22,573 次观看 • 3 个月前

Over the last few months, our engineering teams have been heads down building a high throughput, low-latency blockchain coupled with an exchange replicable across different data centers. We’ve invested thousands of developer hours designing our data infrastructure stack from scratch. We designed our own streaming indexer to serve data to our frontend in real-time, and shipped optimized memory crates to push our general shared IPC to under 100 nanoseconds. This architecture enables us to colocate our sequencers with the key sources of price discovery across all asset classes: Tokyo for crypto, New Jersey for equities, Chicago for commodities. In turn, institutional market makers colocated with GTE bare metal racks will be able to post orders with minimal roundtrip latency, resulting in the tightest spreads and pristine liquidity. We call this GTE Turbo. By optimizing the entire stack from hardware, to the software, into the networking stack, we will build a world where anyone, anywhere, can trade anything, at anytime. lfGTE
0:10

Over the last few months, our engineering teams have been heads down building a high throughput, low-latency blockchain coupled with an exchange replicable across different data centers. We’ve invested thousands of developer hours designing our data infrastructure stack from scratch. We designed our own streaming indexer to serve data to our frontend in real-time, and shipped optimized memory crates to push our general shared IPC to under 100 nanoseconds. This architecture enables us to colocate our sequencers with the key sources of price discovery across all asset classes: Tokyo for crypto, New Jersey for equities, Chicago for commodities. In turn, institutional market makers colocated with GTE bare metal racks will be able to post orders with minimal roundtrip latency, resulting in the tightest spreads and pristine liquidity. We call this GTE Turbo. By optimizing the entire stack from hardware, to the software, into the networking stack, we will build a world where anyone, anywhere, can trade anything, at anytime. lfGTE

Matteo

141,212 次观看 • 9 个月前

We have built a re-entry capsule. And we’re doing a mid-air drop test to demonstrate our dynamics and landing capabilities, a key test before we launch it into space within next 12 months. RSVP to watch us live from Mission Control. We’re creating the OpenAI moment for space, where anyone can develop solutions in space faster and cheaper.
0:14

We have built a re-entry capsule. And we’re doing a mid-air drop test to demonstrate our dynamics and landing capabilities, a key test before we launch it into space within next 12 months. RSVP to watch us live from Mission Control. We’re creating the OpenAI moment for space, where anyone can develop solutions in space faster and cheaper.

Rifath Shaarook

42,012 次观看 • 1 年前

We are releasing AutoResearchExam, a benchmark on open-ended machine learning and engineering tasks. Our benchmark covers seven research areas including model training, data curation, AI safety and interpretability. In each task, we give agents 24 hours with a CPU or GPU machine to develop and improve their solutions through experiments and feedback. We measure both speed and quality with a combined score. Our benchmark has a unique feature: testing if agents create improvements that hold up on data they never see. We find that AI research agents often overfit as they try to improve. We see an interesting head-to-head comparison at the frontier: Astra starts the strongest and holds the lead for up to 19 hours but Fable 5.1 catches up and gets the top performing spot in the final hours. Qwen3.8 Max, Gemini 3.8 Flash and Grok 4.6 all sit on the cost-performance Pareto frontier, giving strong options at lower API budgets. Anthropic's Opus and Fable retain nearly all their validation performance on hidden tests, with gaps of 1.1% and 2.9%. Astra's improvement over Sol extends to generalization too, with that gap falling from 6.9% to 1.7%. (1/n)
0:14

We are releasing AutoResearchExam, a benchmark on open-ended machine learning and engineering tasks. Our benchmark covers seven research areas including model training, data curation, AI safety and interpretability. In each task, we give agents 24 hours with a CPU or GPU machine to develop and improve their solutions through experiments and feedback. We measure both speed and quality with a combined score. Our benchmark has a unique feature: testing if agents create improvements that hold up on data they never see. We find that AI research agents often overfit as they try to improve. We see an interesting head-to-head comparison at the frontier: Astra starts the strongest and holds the lead for up to 19 hours but Fable 5.1 catches up and gets the top performing spot in the final hours. Qwen3.8 Max, Gemini 3.8 Flash and Grok 4.6 all sit on the cost-performance Pareto frontier, giving strong options at lower API budgets. Anthropic's Opus and Fable retain nearly all their validation performance on hidden tests, with gaps of 1.1% and 2.9%. Astra's improvement over Sol extends to generalization too, with that gap falling from 6.9% to 1.7%. (1/n)

Alex Dimakis

1,952,608 次观看 • 6 天前

Sneak peek at some stuff 🫣 This is our local development environment where we can compare what the models generate by default to what they generate with our opinions layered in. So much more to do, but already obvious improvements from just focusing on little details like hanging punctuation, card content alignment, lighting/elevation cues, and edge highlighting of avatars.
0:17

Sneak peek at some stuff 🫣 This is our local development environment where we can compare what the models generate by default to what they generate with our opinions layered in. So much more to do, but already obvious improvements from just focusing on little details like hanging punctuation, card content alignment, lighting/elevation cues, and edge highlighting of avatars.

Adam Wathan

57,471 次观看 • 7 个月前

Help us make Pathfinder even better! The more we talk to our community, the more it validates our belief. Everyone’s favorite shapes are unique, even when they look similar, size makes all the difference! But we want to understand more and to further the science of shape. Our Discord data shows that Xtra Claw, Right Hand, and Wave shapes are among the most popular. Now we want to reach out to mroe people and hear from you! Like ❤️ / Retweet 🔁 this post and complete the survey📋. In one week, we’ll pick a winner 🎁 to be the very first to receive our secret gift box! 🎉
0:13

Help us make Pathfinder even better! The more we talk to our community, the more it validates our belief. Everyone’s favorite shapes are unique, even when they look similar, size makes all the difference! But we want to understand more and to further the science of shape. Our Discord data shows that Xtra Claw, Right Hand, and Wave shapes are among the most popular. Now we want to reach out to mroe people and hear from you! Like ❤️ / Retweet 🔁 this post and complete the survey📋. In one week, we’ll pick a winner 🎁 to be the very first to receive our secret gift box! 🎉

Orbital

11,658 次观看 • 11 个月前