正在加载视频...

视频加载失败

返回首页

Just published a new Flow Matching tutorial (link below). It's a little unorthodox, so I want to hear your feedback before turning it into a video. On the surface, the FM algorithm looks totally arbitrary: 🫤Pick two arbitrary points 🫤(Arbitrarily) decide to draw a straight line 🫤(Arbitrarily) model particle...

juliarturc's profile picture

Julia Turc

17,687 subscribers

17,970 次观看 • 3 个月前 •via X (Twitter)

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

Martin Scorsese once said the reason he wanted to adapt Henry Hill's story into Goodfellas was the way Henry described the arbitrary nature of murder in that world. On a whim, anyone could be killed, it was casual routine. As he explains: “Really it was his description of the way the people related to each other and how incidents would occur. And how if one man did one thing, and the woman did this, how it would have affected a third person. For example, in the killings towards the end of the film, how he began to realize how easy it was for somebody to be killed, just by a word - I think De Niro tells him, "I'm bringing him here tonight." And then he changes his mind, you know - It's all so arbitrary. A lot of it's very arbitrary - there's no big dramatic sit-down in a way.”
1:57

Martin Scorsese once said the reason he wanted to adapt Henry Hill's story into Goodfellas was the way Henry described the arbitrary nature of murder in that world. On a whim, anyone could be killed, it was casual routine. As he explains: “Really it was his description of the way the people related to each other and how incidents would occur. And how if one man did one thing, and the woman did this, how it would have affected a third person. For example, in the killings towards the end of the film, how he began to realize how easy it was for somebody to be killed, just by a word - I think De Niro tells him, "I'm bringing him here tonight." And then he changes his mind, you know - It's all so arbitrary. A lot of it's very arbitrary - there's no big dramatic sit-down in a way.”

Gangster Cinema Central

728,661 次观看 • 2 个月前

With tariffs, we run into the same unsolvable problem that we have with The Federal Reserve. The Fed doesn't know what interest rates should be. It's all arbitrary. In the same way, the president doesn't know what the price of cars, or lumber, or any other product should be. As with the Fed, arbitrary decisions become political decisions. If you lobby well and the administration sees a political benefit, the tariffs will more than likely reflect that. We're further entrenching cronyism into our already corporatist system. What happens when a Democrat is in office? Will the Democrat arbitrarily pick and choose products and tariff rates too? None of this is necessary. But the necessary is being avoided -- the overwhelming size of government, spending, debt and empire. America can never "be great again" with a sprawling military empire, "foreign aid" and constant wars. But as of today, everyone is still refusing to budge. Watch Ron Paul & Chris Rossini below:
23:02

With tariffs, we run into the same unsolvable problem that we have with The Federal Reserve. The Fed doesn't know what interest rates should be. It's all arbitrary. In the same way, the president doesn't know what the price of cars, or lumber, or any other product should be. As with the Fed, arbitrary decisions become political decisions. If you lobby well and the administration sees a political benefit, the tariffs will more than likely reflect that. We're further entrenching cronyism into our already corporatist system. What happens when a Democrat is in office? Will the Democrat arbitrarily pick and choose products and tariff rates too? None of this is necessary. But the necessary is being avoided -- the overwhelming size of government, spending, debt and empire. America can never "be great again" with a sprawling military empire, "foreign aid" and constant wars. But as of today, everyone is still refusing to budge. Watch Ron Paul & Chris Rossini below:

Ron Paul

98,589 次观看 • 1 年前

VIDEO | Delhi: "How is it possible that dead people are in the draft list and living people are on the deleted list? How can a dead person be on a draft list? It can only happen because the BLOs were signing the form themselves...In just two districts of Bihar - Darbhanga and Kaimur, the BLO has marked “not recommended” for 10 to 12%. If we average this for the rest of Bihar, it means 80 lakh people will be marked “not recommended” and subjected to an arbitrary verification. So based on all of this, we can go on till the cows come home. But can I tell you, based on just these five points the CEC mentioned yesterday, this is ludicrous. It is not the job of the Chief Election Commissioners to attack the opposition. Your job, sir, is to go into great detail on the valid queries raised by the opposition", said TMC's Mahua Moitra addressing the INDIA bloc press conference at the Constitution Club.
1:20

VIDEO | Delhi: "How is it possible that dead people are in the draft list and living people are on the deleted list? How can a dead person be on a draft list? It can only happen because the BLOs were signing the form themselves...In just two districts of Bihar - Darbhanga and Kaimur, the BLO has marked “not recommended” for 10 to 12%. If we average this for the rest of Bihar, it means 80 lakh people will be marked “not recommended” and subjected to an arbitrary verification. So based on all of this, we can go on till the cows come home. But can I tell you, based on just these five points the CEC mentioned yesterday, this is ludicrous. It is not the job of the Chief Election Commissioners to attack the opposition. Your job, sir, is to go into great detail on the valid queries raised by the opposition", said TMC's Mahua Moitra addressing the INDIA bloc press conference at the Constitution Club.

Press Trust of India

28,325 次观看 • 11 个月前

i bet you haven't seen a SO101 mounted on a wall like this before if you want to do the same, here is LeSlider: i built it cause i wanted something that can cover my whole desk for tasks like organizing and cleaning i originally wanted to have a belt system like what 3D printers have, but i was too lazy and used a pinion/track with another sts3215 so: > the extra motor shares the same bus as the rest of SO101 > you can have arbitrary length of track > really cheap and easy to assemble and control it turned out better than expected with this, i'm gonna train a model to pick random stuff up across my table and put it into a bin at the end of table (realistically using yolo to scan table, two policies, one for picking up objects, one for dropping)
0:26

i bet you haven't seen a SO101 mounted on a wall like this before if you want to do the same, here is LeSlider: i built it cause i wanted something that can cover my whole desk for tasks like organizing and cleaning i originally wanted to have a belt system like what 3D printers have, but i was too lazy and used a pinion/track with another sts3215 so: > the extra motor shares the same bus as the rest of SO101 > you can have arbitrary length of track > really cheap and easy to assemble and control it turned out better than expected with this, i'm gonna train a model to pick random stuff up across my table and put it into a bin at the end of table (realistically using yolo to scan table, two policies, one for picking up objects, one for dropping)

Binh Pham

98,324 次观看 • 3 个月前

Friends, I want to focus this week on the second reading, which is from the marvelous Letter to the Hebrews. It addresses a very important and very controversial topic—namely, the divine punishment. You would be hard-pressed to say that this is not a motif in the Bible. That’s simply not the case; in fact, it’s a rather major motif. How do we make sense of this theme of divine punishment without falling back into a terrible view of God as an arbitrary, capricious tyrant? This little passage from Hebrews gives us the interpretive key.
14:29

Friends, I want to focus this week on the second reading, which is from the marvelous Letter to the Hebrews. It addresses a very important and very controversial topic—namely, the divine punishment. You would be hard-pressed to say that this is not a motif in the Bible. That’s simply not the case; in fact, it’s a rather major motif. How do we make sense of this theme of divine punishment without falling back into a terrible view of God as an arbitrary, capricious tyrant? This little passage from Hebrews gives us the interpretive key.

Bishop Robert Barron

29,963 次观看 • 11 个月前

PVNR Flyover in Hyderabad brings out some of the worst drivers. Its a 40-60 kmph speed limit. Solid white all over to discourage lane changes. The idea is to get on it and drive in your lane for the entire stretch at or below 60 kmph. NO Overtaking, No Speeding. But some drivers treat it as a racing track- they weave, and cut, and overtake in small gaps putting others at risk. We have to change our racing mindset to achieve safe roads. PS: Unlike highways, flyovers & bridges don't have 'Fast lane' or 'Overtaking lane', everyone must drive below speed limit and follow each other patiently. (I get this response very often so posted this 🫤) #RoadSafety
0:42

PVNR Flyover in Hyderabad brings out some of the worst drivers. Its a 40-60 kmph speed limit. Solid white all over to discourage lane changes. The idea is to get on it and drive in your lane for the entire stretch at or below 60 kmph. NO Overtaking, No Speeding. But some drivers treat it as a racing track- they weave, and cut, and overtake in small gaps putting others at risk. We have to change our racing mindset to achieve safe roads. PS: Unlike highways, flyovers & bridges don't have 'Fast lane' or 'Overtaking lane', everyone must drive below speed limit and follow each other patiently. (I get this response very often so posted this 🫤) #RoadSafety

Naresh

15,013 次观看 • 11 个月前

I know we are all excited about Sora 2, but don't sleep on the other models. Runway Aleph is a powerful tool as you can essentially control AI generated characters using videos recorded with your phone. And it's now on Freepik 🔥 To use Runway Aleph on Freepik, head to the new clip editor. Then pick the Aleph model from the menu (or leave it on Auto). Add a prompt and hit generate. All covered in the short tutorial below.
1:34

I know we are all excited about Sora 2, but don't sleep on the other models. Runway Aleph is a powerful tool as you can essentially control AI generated characters using videos recorded with your phone. And it's now on Freepik 🔥 To use Runway Aleph on Freepik, head to the new clip editor. Then pick the Aleph model from the menu (or leave it on Auto). Add a prompt and hit generate. All covered in the short tutorial below.

Martin LeBlanc

12,050 次观看 • 10 个月前

I’ve run 12+ miles per day for nearly 3,000 straight days and now I’m turning it into a daily series to give you guys full access to how I do it, let you guys pick challenges for me to complete, and hopefully use my story to inspire you guys. And this new series on my running streak is simply called “The Running Steak” (I know really creative!) How should I spice up these runs, what stories do you want to hear about the streak, what do you want me to wear? Leave comments and suggests, anything that is not totally dangerous & impossible is on the table!
1:22

I’ve run 12+ miles per day for nearly 3,000 straight days and now I’m turning it into a daily series to give you guys full access to how I do it, let you guys pick challenges for me to complete, and hopefully use my story to inspire you guys. And this new series on my running streak is simply called “The Running Steak” (I know really creative!) How should I spice up these runs, what stories do you want to hear about the streak, what do you want me to wear? Leave comments and suggests, anything that is not totally dangerous & impossible is on the table!

Greg Peterson

108,767 次观看 • 20 天前

I just created an agentic-workflow to automatically write and publish content for me! It's powered by CrewAI Flows and Llama 3.2, running 100% locally. Tech stack: - CrewAI to build an agentic workflow - FireCrawl for web scraping - Typefully for scheduling Here's how it works: - You provide a link to a website. - It scrapes and saves the data as markdown. - A router triggers the desired Crew of agents. - The Crew prepares a ready-to-publish draft. - Finally, use Typefully to post it to your socials. Totally hands-off and 100% automated! In this video, I provide a deep dive into how it actually works! Find the link to all the code in the next tweet! Enjoy the video! 🥂
6:19

I just created an agentic-workflow to automatically write and publish content for me! It's powered by CrewAI Flows and Llama 3.2, running 100% locally. Tech stack: - CrewAI to build an agentic workflow - FireCrawl for web scraping - Typefully for scheduling Here's how it works: - You provide a link to a website. - It scrapes and saves the data as markdown. - A router triggers the desired Crew of agents. - The Crew prepares a ready-to-publish draft. - Finally, use Typefully to post it to your socials. Totally hands-off and 100% automated! In this video, I provide a deep dive into how it actually works! Find the link to all the code in the next tweet! Enjoy the video! 🥂

Akshay 🚀

98,126 次观看 • 1 年前

Explaining the Ash Nerf confusion: Patchnotes say: "Dash gives less of a boost when not dashing in the direction you're moving" What does that mean ? I dug into it. TLDR: Redirects above 90° are only really possible at normal Slidejump velocities. Angle and speed of your dash will deviate further and further from where you wanna go based on speed and angle change. 90°: Dash seems to work fairly well all the way up to 90° up to a certain velocity. - At Slidejump velocities it works in all directions. - At Superjump velocities you can still dash full 90°. - Above 90° your final direction deviates more and more from your desired direction as the angle increases. - Backwards dash seems to straight up subtract with whatever current velocity you have. Velocity: - As velocity increases even 90° changes are no longer possible. - And as your speed goes above the speed of the dash, dashing backwards will drop you straight down or you'll even continue traveling forward. Nerd yap: I feels like it's not just a straight vector addition but it does in fact have some velocity/angle thresholds in place. My assumption is that >90° changes stop applying the full dash boost at around the velocity of the dash itself so 600-700hu/s. And that above 90° it is a straight up dot product no matter the velocity. You just don't really notice cus of the low initial speed. But players will adjust to it based on feeling and experience anyways so i currently don't care enough to dig deeper🤷‍♂️
0:30

Explaining the Ash Nerf confusion: Patchnotes say: "Dash gives less of a boost when not dashing in the direction you're moving" What does that mean ? I dug into it. TLDR: Redirects above 90° are only really possible at normal Slidejump velocities. Angle and speed of your dash will deviate further and further from where you wanna go based on speed and angle change. 90°: Dash seems to work fairly well all the way up to 90° up to a certain velocity. - At Slidejump velocities it works in all directions. - At Superjump velocities you can still dash full 90°. - Above 90° your final direction deviates more and more from your desired direction as the angle increases. - Backwards dash seems to straight up subtract with whatever current velocity you have. Velocity: - As velocity increases even 90° changes are no longer possible. - And as your speed goes above the speed of the dash, dashing backwards will drop you straight down or you'll even continue traveling forward. Nerd yap: I feels like it's not just a straight vector addition but it does in fact have some velocity/angle thresholds in place. My assumption is that >90° changes stop applying the full dash boost at around the velocity of the dash itself so 600-700hu/s. And that above 90° it is a straight up dot product no matter the velocity. You just don't really notice cus of the low initial speed. But players will adjust to it based on feeling and experience anyways so i currently don't care enough to dig deeper🤷‍♂️

Mokey

67,397 次观看 • 10 个月前

The non-arbitrary era begins today… We’re throwing in our biggest contribution to Bitcoin we could come up with Digital Matter Theory represents an opportunity for all who seek to identify new purpose and value that can be derived from the discoverable patterns within a block We recorded a detailed breakdown of DMT as a concept and explain the operations to formally register the existence of patterns and deploy, mint, and transfer NATs (Non-Arbitrary Tokens) in ▶️ watch here or on YouTube 👇 All of this juice is also in the docs we just published 👇 In there you will find the JSON syntax to mint the very first NAT deployments $NAT $BMT Let the DMT pill consume you 💊 Summary: In this episode, we delve into the intricate universe of Digital Matter Theory (DMT) and its potential applications in the blockchain space. We discuss how DMT enables the creation of non-arbitrary tokens by recognizing patterns in data. Safeguarding the provenance of these elements allows the launch of unique projects tied to them, leading to a diverse ecosystem of tokens. We delve into hybrid tokens, a result of minting NFTs harboring fungible aspects. By parsing through Bitcoin data, meaningful patterns can be codified and used for creating projects. We touch upon the inception of bitmap and the role of communities like Yuga Labs or Rare Sat Society in relating patterns within Bitcoin data to create ecosystem value. We also discuss the concept of 'bit informatics', where the search through Bitcoin's blockchain for patterns holds immense potential for creating new tokens or games. In the process, we unravel the creation and utilization of Nats or Non Arbitrary Tokens. Throughout, we draw parallels between DMT and the early days of NFT, stressing the need for a unified method for creators. We underscore the value potential of non-arbitrary data in the metaverse and the chance to create ecosystems through the recognition of patterns. As with every episode, we invite listener feedback, and encourage questions via YouTube and Twitter. For those wanting to dive deeper into this captivating subject, follow and subscribe for future videos. 🚀Sign up to receive our Ordinal Takeover Newsletter for weekly updates! Disclaimer: The views and opinions expressed by The Block Runner are for informational purposes only and do not constitute financial, investment, or other advice.
55:57

The non-arbitrary era begins today… We’re throwing in our biggest contribution to Bitcoin we could come up with Digital Matter Theory represents an opportunity for all who seek to identify new purpose and value that can be derived from the discoverable patterns within a block We recorded a detailed breakdown of DMT as a concept and explain the operations to formally register the existence of patterns and deploy, mint, and transfer NATs (Non-Arbitrary Tokens) in ▶️ watch here or on YouTube 👇 All of this juice is also in the docs we just published 👇 In there you will find the JSON syntax to mint the very first NAT deployments $NAT $BMT Let the DMT pill consume you 💊 Summary: In this episode, we delve into the intricate universe of Digital Matter Theory (DMT) and its potential applications in the blockchain space. We discuss how DMT enables the creation of non-arbitrary tokens by recognizing patterns in data. Safeguarding the provenance of these elements allows the launch of unique projects tied to them, leading to a diverse ecosystem of tokens. We delve into hybrid tokens, a result of minting NFTs harboring fungible aspects. By parsing through Bitcoin data, meaningful patterns can be codified and used for creating projects. We touch upon the inception of bitmap and the role of communities like Yuga Labs or Rare Sat Society in relating patterns within Bitcoin data to create ecosystem value. We also discuss the concept of 'bit informatics', where the search through Bitcoin's blockchain for patterns holds immense potential for creating new tokens or games. In the process, we unravel the creation and utilization of Nats or Non Arbitrary Tokens. Throughout, we draw parallels between DMT and the early days of NFT, stressing the need for a unified method for creators. We underscore the value potential of non-arbitrary data in the metaverse and the chance to create ecosystems through the recognition of patterns. As with every episode, we invite listener feedback, and encourage questions via YouTube and Twitter. For those wanting to dive deeper into this captivating subject, follow and subscribe for future videos. 🚀Sign up to receive our Ordinal Takeover Newsletter for weekly updates! Disclaimer: The views and opinions expressed by The Block Runner are for informational purposes only and do not constitute financial, investment, or other advice.

ᴛʜᴇ ʙʟᴏᴄᴋ ʀᴜɴɴᴇʀ Podcast 🟧

309,379 次观看 • 2 年前

🎉 INTRODUCING: 🪞 welcome to The Looking Glass: a spatiotemporal ai image + video engine! traverse the history of the world through the eyes of multimodal models; anytime, anywhere, any style 🙌 you know the feeling of standing somewhere old and trying to SEE it? the street before the street, the harbour before the concrete, the hill before the city? The Looking Glass gives you a vehicle. to operate this time-travel-adjacent engine, simply choose a point anywhere on the globe at any point in time (past, present, or future), and pull the lever! it creates an image of what it imagines was happening at that exact spot, in that exact year, at that exact hour, then (optionally) brings it to life as video. it's not a perfect science, but I expect the quality will improve as the prompts get refined and model capabilities progress over time. although the backend boils down to simple API calls, the experience feels akin to multidimensional travel! hallucinated? perhaps! it's about the journey 😁 REPO: 🐉 one page. no backend. no account. multiple model options. runs on OpenRouter. your key, your browser, your archive. and as always, totally free to use and open source under AGPL-3.0 🤗 hope you enjoy the demo video below (sped up 2x cause I know how attention spans be these days) EX LOCO, PER VITRUM, AD OMNE TEMPUS from place, through glass, to all time ⊰-•-•✧•-•-⦑/L\O/V\E/,\P/L\I/N\Y/⦒-•-•✧•-•-⊱
1:12

🎉 INTRODUCING: 🪞 welcome to The Looking Glass: a spatiotemporal ai image + video engine! traverse the history of the world through the eyes of multimodal models; anytime, anywhere, any style 🙌 you know the feeling of standing somewhere old and trying to SEE it? the street before the street, the harbour before the concrete, the hill before the city? The Looking Glass gives you a vehicle. to operate this time-travel-adjacent engine, simply choose a point anywhere on the globe at any point in time (past, present, or future), and pull the lever! it creates an image of what it imagines was happening at that exact spot, in that exact year, at that exact hour, then (optionally) brings it to life as video. it's not a perfect science, but I expect the quality will improve as the prompts get refined and model capabilities progress over time. although the backend boils down to simple API calls, the experience feels akin to multidimensional travel! hallucinated? perhaps! it's about the journey 😁 REPO: 🐉 one page. no backend. no account. multiple model options. runs on OpenRouter. your key, your browser, your archive. and as always, totally free to use and open source under AGPL-3.0 🤗 hope you enjoy the demo video below (sped up 2x cause I know how attention spans be these days) EX LOCO, PER VITRUM, AD OMNE TEMPUS from place, through glass, to all time ⊰-•-•✧•-•-⦑/L\O/V\E/,\P/L\I/N\Y/⦒-•-•✧•-•-⊱

Pliny the Liberator 🐉󠅫󠄼󠄿󠅆󠄵󠄐󠅀󠄼󠄹󠄾󠅉󠅭

233,115 次观看 • 1 天前

One of the longest development periods for a AAA video game ever. ⏳ Beyond Good & Evil 2 has been in development since the 2008 teaser, with multiple restarts and reboots along the way. The last real footage we saw was back in 2018. It's believed the current incarnation has been underway since 2017. Somehow it's still not cancelled. It even survived Ubisoft's recent wave of project cuts. And the company just confirmed it's still a priority for their open-world adventure strategy. It will be a prequel to the first game, set in the early years of characters like Jade. We've also heard it supports single-player (fully playable solo) and multiplayer/co-op modes in a shared universe. There was a report circulating in 2022 that according to devs, the game was still in pre-production at that time. At this point, even if it comes out and it's a success, it's hard to see how it turns a profit after all the time and resources they've probably spent on it... 🫤
0:58

One of the longest development periods for a AAA video game ever. ⏳ Beyond Good & Evil 2 has been in development since the 2008 teaser, with multiple restarts and reboots along the way. The last real footage we saw was back in 2018. It's believed the current incarnation has been underway since 2017. Somehow it's still not cancelled. It even survived Ubisoft's recent wave of project cuts. And the company just confirmed it's still a priority for their open-world adventure strategy. It will be a prequel to the first game, set in the early years of characters like Jade. We've also heard it supports single-player (fully playable solo) and multiplayer/co-op modes in a shared universe. There was a report circulating in 2022 that according to devs, the game was still in pre-production at that time. At this point, even if it comes out and it's a success, it's hard to see how it turns a profit after all the time and resources they've probably spent on it... 🫤

𝑨𝒔𝒉𝒆𝒏 𝑶𝒏𝒆

24,926 次观看 • 6 个月前

Standard Intelligence's devansh responds to roon's tweet that "text is the universal interface," and explains why their new foundation model is trained on video: "At some point in the arbitrarily long future, if we only use text models, we could force most things to be text. But I think there are just a lot of things that are much more native when done from a computer-use [perspective]." "GUIs are designed for humans to use. We have this massive long tail of things on the internet that are entirely undoable by LLMs." "For example, when I do ML engineering most of my time is spent doing the grunt work of engineering. It's a lot of looking at graphs, analyzing, and comparing loss curves. You can do this in text, but it's a much larger pain than doing it in the native interface." "There's a reason humans don't interact with a computer purely through text, it would kind of suck."
1:47

Standard Intelligence's devansh responds to roon's tweet that "text is the universal interface," and explains why their new foundation model is trained on video: "At some point in the arbitrarily long future, if we only use text models, we could force most things to be text. But I think there are just a lot of things that are much more native when done from a computer-use [perspective]." "GUIs are designed for humans to use. We have this massive long tail of things on the internet that are entirely undoable by LLMs." "For example, when I do ML engineering most of my time is spent doing the grunt work of engineering. It's a lot of looking at graphs, analyzing, and comparing loss curves. You can do this in text, but it's a much larger pain than doing it in the native interface." "There's a reason humans don't interact with a computer purely through text, it would kind of suck."

TBPN

64,434 次观看 • 5 个月前

Moscow cynically claimed that the right for self-defense without taking the UN Charter into account leads to chaos. Commenting on the escalation between Israel and Iran, Lavrov said that arbitrary exercise of the right to self-defense without respect for the norms of the UN Charter undermines the international order: If each country is allowed to decide for itself when to apply this right without looking back at the UN Charter, it will not be a world order, but complete chaos, he said. Oh really? And what is Russia doing in Ukraine? You're the ones who destroyed all thr systems, the whole order and plunged the world into chaos.
1:00

Moscow cynically claimed that the right for self-defense without taking the UN Charter into account leads to chaos. Commenting on the escalation between Israel and Iran, Lavrov said that arbitrary exercise of the right to self-defense without respect for the norms of the UN Charter undermines the international order: If each country is allowed to decide for itself when to apply this right without looking back at the UN Charter, it will not be a world order, but complete chaos, he said. Oh really? And what is Russia doing in Ukraine? You're the ones who destroyed all thr systems, the whole order and plunged the world into chaos.

Anton Gerashchenko

73,906 次观看 • 1 年前

This is a pretty wild model! You can use it to turn an image into a 3D object with texture. The quality is out of this world! I'm not even a designer, and I've been using this nonstop for the last 2 hours. The model is Hunyuan 3D 2.1. It's open source. You'll find model weights, training/inference code, data pipelines, and architecture on their repository. You can even fine-tune it if you want! GitHub Repository: By the way, the model runs on consumer-grade GPUs. You don't need a datacenter for this! I've been using the model from the HuggingFace demo page: To use it, go to the link and upload an image. That's it! Check out the video I recorded for a couple of examples.
2:42

This is a pretty wild model! You can use it to turn an image into a 3D object with texture. The quality is out of this world! I'm not even a designer, and I've been using this nonstop for the last 2 hours. The model is Hunyuan 3D 2.1. It's open source. You'll find model weights, training/inference code, data pipelines, and architecture on their repository. You can even fine-tune it if you want! GitHub Repository: By the way, the model runs on consumer-grade GPUs. You don't need a datacenter for this! I've been using the model from the HuggingFace demo page: To use it, go to the link and upload an image. That's it! Check out the video I recorded for a couple of examples.

Santiago

44,783 次观看 • 1 年前

iit's wild that this still isn't the default way people use AI. Andrew Ng, a co-founder of Google Brain, gave a talk that crossed 2 million views: stop feeding your AI a folder of notes, build it a graph it can actually think with. his point is simple, stop making your AI read the whole vault to answer one thing. give it a graph, a router that points, an index that scores, edges that link, and it walks straight to the answer. it compounds, cheaper and faster the more you add. here's the whole thing: > point Claude at your notes and have it write a tiny router, your CLAUDE.md > build an index, one line per note, so it always knows where an answer lives > keep each note to one idea, and link the ones that depend on each other > make retrieval run as code, so it scores and opens the right files before the model thinks > tell it to prove the graph beats a flat folder on speed and cost, then keep the index fresh on every write set it up once and your AI stops guessing. it already knows your world before you type a word. full 12-step guide below. bookmark this.
0:23

iit's wild that this still isn't the default way people use AI. Andrew Ng, a co-founder of Google Brain, gave a talk that crossed 2 million views: stop feeding your AI a folder of notes, build it a graph it can actually think with. his point is simple, stop making your AI read the whole vault to answer one thing. give it a graph, a router that points, an index that scores, edges that link, and it walks straight to the answer. it compounds, cheaper and faster the more you add. here's the whole thing: > point Claude at your notes and have it write a tiny router, your CLAUDE.md > build an index, one line per note, so it always knows where an answer lives > keep each note to one idea, and link the ones that depend on each other > make retrieval run as code, so it scores and opens the right files before the model thinks > tell it to prove the graph beats a flat folder on speed and cost, then keep the index fresh on every write set it up once and your AI stops guessing. it already knows your world before you type a word. full 12-step guide below. bookmark this.

KingWilliam

15,099 次观看 • 8 天前

Sergio Garcia reacts to LIV Golf’s OWGR approval: “It's definitely a step forward. Is it fair? I mean, I guess time will tell us. It doesn't feel like it's totally fair. With time, obviously. Now the first few weeks obviously when one of us is winning, that guy is going to make a jump in the rankings, which is great, but then every time you finish 11th or worst you're getting a zero and you're getting an extra event on your divisor. So that in the near, a little farther future, it can hurt a lot. But like I said, it's a step forward, so it is positive. I think that at the end of the day, it's going to help some of these kids performing well to move up the rankings. Hopefully they play really well. “But unfortunately only 10 guys can finish in the top 10. Well, unless there's a couple of ties. “It would be nice if it was expanded a little bit farther as we probably think it should. But hopefully it'll come soon.”
1:21

Sergio Garcia reacts to LIV Golf’s OWGR approval: “It's definitely a step forward. Is it fair? I mean, I guess time will tell us. It doesn't feel like it's totally fair. With time, obviously. Now the first few weeks obviously when one of us is winning, that guy is going to make a jump in the rankings, which is great, but then every time you finish 11th or worst you're getting a zero and you're getting an extra event on your divisor. So that in the near, a little farther future, it can hurt a lot. But like I said, it's a step forward, so it is positive. I think that at the end of the day, it's going to help some of these kids performing well to move up the rankings. Hopefully they play really well. “But unfortunately only 10 guys can finish in the top 10. Well, unless there's a couple of ties. “It would be nice if it was expanded a little bit farther as we probably think it should. But hopefully it'll come soon.”

Flushing It

28,934 次观看 • 5 个月前