Загрузка видео...

Не удалось загрузить видео

На главную

Throughout my journey in developing multimodal models, I’ve always wanted a framework that lets me plug & play modality encoders/decoders on top of an auto-regressive LLM. I want to prototype fast, try new architectures, and have my demo files scale effortlessly — with full support for parallelism and optimization....

Brian_Bo_Li's profile picture

Brian Li

3,025 subscribers

54,840 просмотров • 10 месяцев назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

A major question in multimodal modeling is how to leverage strong pre-trained models, such as industry-scale LLMs and VLMs, during training to avoid starting from scratch. This is important as the demand for rich and domain-specific multimodal models continues to increase, while training them from scratch is obviously impractical due to limited data and compute. The so-called “any-to-any” multimodal models can model a large diverse dictionary of modalities. That’s good. Their downside is that their architectures are often not decoder-only, which has limited their performance in practice and prevents them from leveraging strong pre-trained decoder-only models as priors. We are releasing MODUS (ICML Conference '26) a decoder-only any-to-any multimodal model to address some of these questions. A single transformer decoder predicts any modality from any others with no modality-specific heads, losses, or task pipelines. We show efficient adaptation of established models, e.g., BAGEL, to rich any-to-any multimodal modeling. We are releasing 14B to 77B parameter models. All materials are open-source. The download links and demos here 🧵
0:18

A major question in multimodal modeling is how to leverage strong pre-trained models, such as industry-scale LLMs and VLMs, during training to avoid starting from scratch. This is important as the demand for rich and domain-specific multimodal models continues to increase, while training them from scratch is obviously impractical due to limited data and compute. The so-called “any-to-any” multimodal models can model a large diverse dictionary of modalities. That’s good. Their downside is that their architectures are often not decoder-only, which has limited their performance in practice and prevents them from leveraging strong pre-trained decoder-only models as priors. We are releasing MODUS (ICML Conference '26) a decoder-only any-to-any multimodal model to address some of these questions. A single transformer decoder predicts any modality from any others with no modality-specific heads, losses, or task pipelines. We show efficient adaptation of established models, e.g., BAGEL, to rich any-to-any multimodal modeling. We are releasing 14B to 77B parameter models. All materials are open-source. The download links and demos here 🧵

Amir Zamir

24,028 просмотров • 2 месяцев назад

We are entering an extremely exciting era for open-weight models. Kimi K2.6 now feels like a top agentic model. I took it for a spin via Fireworks AI fast inference APIs. Kimi K2.6 has impressive agentic capabilities, design skills, and the ability to synthesize large amounts of information. I built a little Skill that produces survey papers on any AI research topic you want. (see example in the clip) You can use the skill to tell your agent to generate a survey on whatever topic and watch it go to work. The artifact was fully generated by Kimi.ai's Kimi K2.6. It's cheap and fast. Next step for me is to explore ways to continue integrating the capabilities of these models on use cases like automating my LLM knowledge bases and augmenting my agent memory capabilities. Stay tuned for more.
0:19

We are entering an extremely exciting era for open-weight models. Kimi K2.6 now feels like a top agentic model. I took it for a spin via Fireworks AI fast inference APIs. Kimi K2.6 has impressive agentic capabilities, design skills, and the ability to synthesize large amounts of information. I built a little Skill that produces survey papers on any AI research topic you want. (see example in the clip) You can use the skill to tell your agent to generate a survey on whatever topic and watch it go to work. The artifact was fully generated by Kimi.ai's Kimi K2.6. It's cheap and fast. Next step for me is to explore ways to continue integrating the capabilities of these models on use cases like automating my LLM knowledge bases and augmenting my agent memory capabilities. Stay tuned for more.

elvis

47,678 просмотров • 5 месяцев назад

This app has racked 500K installs using this UGC format. This is one of the best app growth strategies I’ve seen by this language fluency ai app. The video got 9.9M views 990K likes The app is doing: $500K MRR $6M ARR With formats like: “why tf is English so hard 🥀” …then seamlessly transitions into the app demo. I studied and reverse-engineered all of their UGC and content marketing strategies. Then I built a complete UGC system that lets you replicate the same growth framework, create viral concepts consistently, and scale your app much faster without guessing what to post. If you want to know how they grew so quickly and replicate their success… Comment “VIRAL” and I’ll send the full guide (must be following so I can DM).
0:13

This app has racked 500K installs using this UGC format. This is one of the best app growth strategies I’ve seen by this language fluency ai app. The video got 9.9M views 990K likes The app is doing: $500K MRR $6M ARR With formats like: “why tf is English so hard 🥀” …then seamlessly transitions into the app demo. I studied and reverse-engineered all of their UGC and content marketing strategies. Then I built a complete UGC system that lets you replicate the same growth framework, create viral concepts consistently, and scale your app much faster without guessing what to post. If you want to know how they grew so quickly and replicate their success… Comment “VIRAL” and I’ll send the full guide (must be following so I can DM).

Daniel

16,811 просмотров • 2 месяцев назад

We’re excited to introduce ShinkaEvolve: An open-source framework that evolves programs for scientific discovery with unprecedented sample-efficiency. Blog: Code: Like AlphaEvolve and its variants, our framework leverages LLMs to find state-of-the-art solutions to complex problems, but using orders of magnitude fewer resources! Many evolutionary AI systems are powerful but act like brute-force engines, burning thousands of samples to find good solutions. This makes discovery slow and expensive. We took inspiration from the efficiency of nature. ‘Shinka’ (進化) is Japanese for evolution, and we designed our system to be just as resourceful. On the classic circle packing optimization problem, ShinkaEvolve discovered a new state-of-the-art solution using only 150 samples. This is a big leap in efficiency compared to previous methods that required thousands of evaluations. We applied ShinkaEvolve to a diverse set of hard problems with real-world applications: 1/ AIME Math Reasoning: It evolved sophisticated agentic scaffolds that significantly outperform strong baselines, discovering an entire Pareto frontier of solutions trading performance for efficiency. 2/ Competitive Programming: On ALE-Bench (a benchmark for NP-Hard optimization problems), ShinkaEvolve took the best existing agent's solutions and improved them, turning a 5th place solution on one task into a 2nd place leaderboard rank in a competitive programming competition. 3/ LLM Training: We even turned ShinkaEvolve inward to improve LLMs themselves. It tackled the open challenge of designing load balancing losses for Mixture-of-Experts (MoE) models. It discovered a novel loss function that leads to better expert specialization and consistently improves model performance and perplexity. ShinkaEvolve achieves its remarkable sample-efficiency through three key innovations that work together: (1) an adaptive parent sampling strategy to balance exploration and exploitation, (2) novelty-based rejection filtering to avoid redundant work, and (3) a bandit-based LLM ensemble that dynamically picks the best model for the job. By making ShinkaEvolve open-source and highly sample-efficient, our goal is to democratize access to advanced, open-ended discovery tools. Our vision for ShinkaEvolve is to be an easy-to-use companion tool to help scientists and engineers with their daily work. We believe that building more efficient, nature-inspired systems is key to unlocking the future of AI-driven scientific research. We are excited to see what the community builds with it! Learn more in our technical report:
0:11

We’re excited to introduce ShinkaEvolve: An open-source framework that evolves programs for scientific discovery with unprecedented sample-efficiency. Blog: Code: Like AlphaEvolve and its variants, our framework leverages LLMs to find state-of-the-art solutions to complex problems, but using orders of magnitude fewer resources! Many evolutionary AI systems are powerful but act like brute-force engines, burning thousands of samples to find good solutions. This makes discovery slow and expensive. We took inspiration from the efficiency of nature. ‘Shinka’ (進化) is Japanese for evolution, and we designed our system to be just as resourceful. On the classic circle packing optimization problem, ShinkaEvolve discovered a new state-of-the-art solution using only 150 samples. This is a big leap in efficiency compared to previous methods that required thousands of evaluations. We applied ShinkaEvolve to a diverse set of hard problems with real-world applications: 1/ AIME Math Reasoning: It evolved sophisticated agentic scaffolds that significantly outperform strong baselines, discovering an entire Pareto frontier of solutions trading performance for efficiency. 2/ Competitive Programming: On ALE-Bench (a benchmark for NP-Hard optimization problems), ShinkaEvolve took the best existing agent's solutions and improved them, turning a 5th place solution on one task into a 2nd place leaderboard rank in a competitive programming competition. 3/ LLM Training: We even turned ShinkaEvolve inward to improve LLMs themselves. It tackled the open challenge of designing load balancing losses for Mixture-of-Experts (MoE) models. It discovered a novel loss function that leads to better expert specialization and consistently improves model performance and perplexity. ShinkaEvolve achieves its remarkable sample-efficiency through three key innovations that work together: (1) an adaptive parent sampling strategy to balance exploration and exploitation, (2) novelty-based rejection filtering to avoid redundant work, and (3) a bandit-based LLM ensemble that dynamically picks the best model for the job. By making ShinkaEvolve open-source and highly sample-efficient, our goal is to democratize access to advanced, open-ended discovery tools. Our vision for ShinkaEvolve is to be an easy-to-use companion tool to help scientists and engineers with their daily work. We believe that building more efficient, nature-inspired systems is key to unlocking the future of AI-driven scientific research. We are excited to see what the community builds with it! Learn more in our technical report:

Sakana AI

360,407 просмотров • 11 месяцев назад

I just built my own wiki generator plugin for my agents. My agents can now generate wikis for anything I ask. One of my favorite wikis is called PaperWiki. This is a great example of what Andrej Karpathy describes. It uses obsidian vaults to organize papers, retrieve LLM-generated summaries, diagrams, and other advanced views for paper exploration. When Obsidian UI is not enough, I use my own artifact generator inside my agent orchestrator (see clip for example). This allows my agents to build any kind of view or exploration feature that I need. The papers are all curated with automations and several rules/patterns I have manually built over the years. On the surface, this looks basic. But behind the scenes, there are advanced search capabilities, connections, metadata, derived data, and other interesting bits of information that are extremely useful for my research agents. This is mostly built for agents. The artifact preview is just a high-level way to validate and quickly assess the quality of the wiki, suggest improvements, and it's also great for research. I use tobi lutke's qmd for all search capabilities. Everything is markdown. The summaries and even the diagrams. The wiki updates on its own based on several automations I have optimized over the past couple of weeks. The wiki grows and self-improves based on several requirements important for my research use cases. This is as personalized as it gets. There is nothing like it out there. And I use my research expertise to continue improving it over time. This is a vanilla wiki. There are so many things I want to build on top of this. Different aggregations, views, artifacts, etc. All to help automate more of my research work and accelerate productivity. I think the biggest leverage here is how powerful this could be for discovery and experimentation. One of my goals is to use it to find deeper connections and insights that would otherwise elude the top human researchers and use those to generate interesting new hypotheses and research experiments. That way, my agents can use autoresearch to explore research ideas at the frontier. Stay tuned for more.
0:16

I just built my own wiki generator plugin for my agents. My agents can now generate wikis for anything I ask. One of my favorite wikis is called PaperWiki. This is a great example of what Andrej Karpathy describes. It uses obsidian vaults to organize papers, retrieve LLM-generated summaries, diagrams, and other advanced views for paper exploration. When Obsidian UI is not enough, I use my own artifact generator inside my agent orchestrator (see clip for example). This allows my agents to build any kind of view or exploration feature that I need. The papers are all curated with automations and several rules/patterns I have manually built over the years. On the surface, this looks basic. But behind the scenes, there are advanced search capabilities, connections, metadata, derived data, and other interesting bits of information that are extremely useful for my research agents. This is mostly built for agents. The artifact preview is just a high-level way to validate and quickly assess the quality of the wiki, suggest improvements, and it's also great for research. I use tobi lutke's qmd for all search capabilities. Everything is markdown. The summaries and even the diagrams. The wiki updates on its own based on several automations I have optimized over the past couple of weeks. The wiki grows and self-improves based on several requirements important for my research use cases. This is as personalized as it gets. There is nothing like it out there. And I use my research expertise to continue improving it over time. This is a vanilla wiki. There are so many things I want to build on top of this. Different aggregations, views, artifacts, etc. All to help automate more of my research work and accelerate productivity. I think the biggest leverage here is how powerful this could be for discovery and experimentation. One of my goals is to use it to find deeper connections and insights that would otherwise elude the top human researchers and use those to generate interesting new hypotheses and research experiments. That way, my agents can use autoresearch to explore research ideas at the frontier. Stay tuned for more.

elvis

67,308 просмотров • 5 месяцев назад

I am extremely excited to announce my verbal commitment to play D1 soccer at the University of Florida. I want to thank God for blessing me with this amazing opportunity. I would also like to thank my parents for their dedication and commitment along this journey. I truly could not have achieved this without their unwavering support and sacrifices. Thank you to all of the coaches and teammates that have helped me along the way, challenging me and shaping me into the player that I am today. I would also like to give a huge thank you to Coach Samantha, Adam, Austin, Kelly, and Kerri for giving me this opportunity and believing in me. I am so grateful to be a Gator!🐊 Gators Soccer Samantha Bohon Adam Godwin Austin O'Connor Kelly Altman Katelyn Kight Kerri Hanks Katelyn Kight em Lauren Donovan Ben Chase Abby Gemma
0:11

I am extremely excited to announce my verbal commitment to play D1 soccer at the University of Florida. I want to thank God for blessing me with this amazing opportunity. I would also like to thank my parents for their dedication and commitment along this journey. I truly could not have achieved this without their unwavering support and sacrifices. Thank you to all of the coaches and teammates that have helped me along the way, challenging me and shaping me into the player that I am today. I would also like to give a huge thank you to Coach Samantha, Adam, Austin, Kelly, and Kerri for giving me this opportunity and believing in me. I am so grateful to be a Gator!🐊 Gators Soccer Samantha Bohon Adam Godwin Austin O'Connor Kelly Altman Katelyn Kight Kerri Hanks Katelyn Kight em Lauren Donovan Ben Chase Abby Gemma

Cali Smith ECNL 2027

31,895 просмотров • 1 год назад

IMPORANT ANNOUNCMENT! ⊹₊˚‧︵‿₊୨ᰔ୧₊‿︵‧˚₊⊹ i want to make an official post after all the stuff that blew up and went down since yesterday and update everyone, first of all i want to thank each and every one of you that have sent me sweet messages of support, i see them all and i hear you! sadly there is SO many- i cant like and reply to them all, so please know im so grateful for you guys who have said something nice! i am also safe and have no bad thoughts currently, so please do not worry too much about me! regarding streaming and content creation, i really thought about the beginning of my passion for content creation. it started out as little ol me sitting on my ipad watching youtube for hours wanting to be just like my favorite youtubers. why didnt i start off on youtube? i saw streaming was a huge hit during the pandemic and started preparing myself on Twitch, following whatever was big at the time. after going through so much hardships with streaming, i have decided to go back to my original love for youtube and now i'll make pre-recorded content as well as shorts on other platforms. i may stream for fun occasionally but i am totally changing it up for 2026! i hope everyone can support my decision and its easier for everyone to consume my content on their own time rather than scheduled streams. and again, thank you for 13K+ followers and all the love- it truly makes my heart warm inside. you guys are all stuck with this crazy snow leopard forever, i dont make the rules! /ᐠ˵- ⩊ -˵マ TLDR : thanking everyone for being kind and i am now moving from streaming to making pre-recorded content for easier viewing and better schedule ♡
0:12

IMPORANT ANNOUNCMENT! ⊹₊˚‧︵‿₊୨ᰔ୧₊‿︵‧˚₊⊹ i want to make an official post after all the stuff that blew up and went down since yesterday and update everyone, first of all i want to thank each and every one of you that have sent me sweet messages of support, i see them all and i hear you! sadly there is SO many- i cant like and reply to them all, so please know im so grateful for you guys who have said something nice! i am also safe and have no bad thoughts currently, so please do not worry too much about me! regarding streaming and content creation, i really thought about the beginning of my passion for content creation. it started out as little ol me sitting on my ipad watching youtube for hours wanting to be just like my favorite youtubers. why didnt i start off on youtube? i saw streaming was a huge hit during the pandemic and started preparing myself on Twitch, following whatever was big at the time. after going through so much hardships with streaming, i have decided to go back to my original love for youtube and now i'll make pre-recorded content as well as shorts on other platforms. i may stream for fun occasionally but i am totally changing it up for 2026! i hope everyone can support my decision and its easier for everyone to consume my content on their own time rather than scheduled streams. and again, thank you for 13K+ followers and all the love- it truly makes my heart warm inside. you guys are all stuck with this crazy snow leopard forever, i dont make the rules! /ᐠ˵- ⩊ -˵マ TLDR : thanking everyone for being kind and i am now moving from streaming to making pre-recorded content for easier viewing and better schedule ♡

Snowie ❄️🐾 Snow Leppy VTuber

30,178 просмотров • 8 месяцев назад

My dual RTX PRO 6000 setup is currently training a Draft model for Qwen 3.6 27B! 🔥 I'm taking the paper DeepSeek dropped on 6/26 and going for a super ambitious application to the 27B scale. Thanks to my homelab, I was able to dive straight in — I read the paper and immediately started experimenting. The amount I've learned has been insane: - How memory bandwidth bottlenecks speed and clever ways to hack around it - Methods to train the draft model and boost its accuracy - Mechanisms to reference tokens all the way back to the previous one to skyrocket draft acceptance rates - The impact of Attention vs. GateDeltaNet on speculative decoding performance and how to handle those differences - The unique approaches and trade-offs of MTP, Dflash, JetSpec, and DSpark I could go on forever, but just from speculative decoding alone I've learned so much. The 27B architecture feels way more DSpark-native than JetSpec, so once draft training finishes, I'm going all-in with DSpark! My goal is to beat existing speculative decoding speeds outright — no task-specific shortcuts or cheating, pure general improvement. If you're into this kind of research, I'd love to hear your thoughts, impressions, and any suggestions — please reply! 🚀
0:17

My dual RTX PRO 6000 setup is currently training a Draft model for Qwen 3.6 27B! 🔥 I'm taking the paper DeepSeek dropped on 6/26 and going for a super ambitious application to the 27B scale. Thanks to my homelab, I was able to dive straight in — I read the paper and immediately started experimenting. The amount I've learned has been insane: - How memory bandwidth bottlenecks speed and clever ways to hack around it - Methods to train the draft model and boost its accuracy - Mechanisms to reference tokens all the way back to the previous one to skyrocket draft acceptance rates - The impact of Attention vs. GateDeltaNet on speculative decoding performance and how to handle those differences - The unique approaches and trade-offs of MTP, Dflash, JetSpec, and DSpark I could go on forever, but just from speculative decoding alone I've learned so much. The 27B architecture feels way more DSpark-native than JetSpec, so once draft training finishes, I'm going all-in with DSpark! My goal is to beat existing speculative decoding speeds outright — no task-specific shortcuts or cheating, pure general improvement. If you're into this kind of research, I'd love to hear your thoughts, impressions, and any suggestions — please reply! 🚀

Hikari∣LocalLLM⚡

56,870 просмотров • 2 месяцев назад

🚨 YOUR GPU IS PROBABLY WASTING MORE THAN YOU THINK. vLLM is built to squeeze far more useful work out of your GPU when serving LLMs. Running an LLM at scale isn’t just about having a powerful GPU. The real problem is how efficiently you use its memory and compute. That’s where vLLM comes in. → High-throughput LLM inference and serving → PagedAttention for smarter KV-cache memory management → Continuous batching to keep GPUs busy → Prefix caching + chunked prefill → OpenAI-compatible API out of the box → Supports a huge range of modern LLM architectures → Quantization support for running models more efficiently And the crazy part? You can start an OpenAI-compatible inference server with: `vllm serve ` So your application can talk to your own model almost like it’s talking to OpenAI. The bigger idea: Don’t just buy more GPUs. Make the GPUs you already have work harder. That’s why vLLM has become such a major project in LLM inference. 🔥 #vLLM #AI #LLM #Inference #GPU #MachineLearning #AIInfrastructure #OpenSource #AIAgents #Developers
0:20

🚨 YOUR GPU IS PROBABLY WASTING MORE THAN YOU THINK. vLLM is built to squeeze far more useful work out of your GPU when serving LLMs. Running an LLM at scale isn’t just about having a powerful GPU. The real problem is how efficiently you use its memory and compute. That’s where vLLM comes in. → High-throughput LLM inference and serving → PagedAttention for smarter KV-cache memory management → Continuous batching to keep GPUs busy → Prefix caching + chunked prefill → OpenAI-compatible API out of the box → Supports a huge range of modern LLM architectures → Quantization support for running models more efficiently And the crazy part? You can start an OpenAI-compatible inference server with: `vllm serve ` So your application can talk to your own model almost like it’s talking to OpenAI. The bigger idea: Don’t just buy more GPUs. Make the GPUs you already have work harder. That’s why vLLM has become such a major project in LLM inference. 🔥 #vLLM #AI #LLM #Inference #GPU #MachineLearning #AIInfrastructure #OpenSource #AIAgents #Developers

Vikas gupta

14,138 просмотров • 8 дней назад

LLM Artifacts Connected to Andrej Karpathy's LLM Knowledge base idea, I've been building out a fun way to generate dynamic artifacts from these knowledge bases with the goal of discovering and revealing meaningful and deeper insights. LLM KBs are hard to consume for humans, as I think they are more built for agents. So the question is, what form would be useful for humans to take actions and make important decisions? That's what I am trying to figure out with these artifacts. The artifact example shows a pulse on HN discussions around AI-related stories. The insights can go deeper, of course, but this is already super fun and thought-provoking, like some of my favorite podcasts. The format and depth matter a lot. The aggregation skills of agents are outstanding if you tune the prompts and skill carefully. I built this artifact generator in a few minutes through an agent skill, but I feel like there are so many ways that LLM-generated information can be used and consumed. Like generating deeper insights and analysis, and things that are just not feasible for humans today. The generated artifact (including its data and design) serves as reusable templates or can be updated in real-time via auomations, which is something I am also working on. It is truly an insane way to monitor and track information. Better than a newsletter. Better than newspapers. There is something about this that gets me really excited about the future of AI agents for knowledge generation and discovery. Lots of hidden gems everywhere just waiting to be discovered and acted on if the information is presented correctly. This is not perfect. The format, style/prose can be improved, but this is easy to customize via skill. You can personalize it to your liking. I feel like these dynamic artifacts are going to emerge as a strong new medium to stay on the cutting edge of things, both for agents and humans. My target is research, of course. This was just a basic example. Besides animation, I am also targeting other components like voice, videos, images, slides, etc. This space is full of opportunities to explore. Skill for this coming soon.
0:18

LLM Artifacts Connected to Andrej Karpathy's LLM Knowledge base idea, I've been building out a fun way to generate dynamic artifacts from these knowledge bases with the goal of discovering and revealing meaningful and deeper insights. LLM KBs are hard to consume for humans, as I think they are more built for agents. So the question is, what form would be useful for humans to take actions and make important decisions? That's what I am trying to figure out with these artifacts. The artifact example shows a pulse on HN discussions around AI-related stories. The insights can go deeper, of course, but this is already super fun and thought-provoking, like some of my favorite podcasts. The format and depth matter a lot. The aggregation skills of agents are outstanding if you tune the prompts and skill carefully. I built this artifact generator in a few minutes through an agent skill, but I feel like there are so many ways that LLM-generated information can be used and consumed. Like generating deeper insights and analysis, and things that are just not feasible for humans today. The generated artifact (including its data and design) serves as reusable templates or can be updated in real-time via auomations, which is something I am also working on. It is truly an insane way to monitor and track information. Better than a newsletter. Better than newspapers. There is something about this that gets me really excited about the future of AI agents for knowledge generation and discovery. Lots of hidden gems everywhere just waiting to be discovered and acted on if the information is presented correctly. This is not perfect. The format, style/prose can be improved, but this is easy to customize via skill. You can personalize it to your liking. I feel like these dynamic artifacts are going to emerge as a strong new medium to stay on the cutting edge of things, both for agents and humans. My target is research, of course. This was just a basic example. Besides animation, I am also targeting other components like voice, videos, images, slides, etc. This space is full of opportunities to explore. Skill for this coming soon.

elvis

31,314 просмотров • 5 месяцев назад

I’ve been craving a video editor with motion baked in, designed to work with my own AI agents. Claude, Codex, whatever I want to use. I don’t want to be blocked by someone else’s credit system for every little interaction. I also wanted a Figma-like canvas/editor for when I want to jump in and tweak any little detail. So we built it: Video & audio editing: cut, zoom, speed up, add B-roll. Motion, animation, 3D, custom shaders & effects, custom 3D models, dynamic components & templates, audio and beat matching. And it’s all programmable. Design your own assets directly on the canvas, bring them in from Figma or the web, or just let the agent find them, mock them up, or record what it needs automatically. You can even drop in your screen recordings and just let loose. Supercut users: yes, it’ll have first-class support for your recordings, but it’ll work with anything. I’m going to be dropping a lot more examples and tutorials, so follow along. DM me for early access if you’re willing to give feedback. We’ll open it up to everyone very soon.
0:19

I’ve been craving a video editor with motion baked in, designed to work with my own AI agents. Claude, Codex, whatever I want to use. I don’t want to be blocked by someone else’s credit system for every little interaction. I also wanted a Figma-like canvas/editor for when I want to jump in and tweak any little detail. So we built it: Video & audio editing: cut, zoom, speed up, add B-roll. Motion, animation, 3D, custom shaders & effects, custom 3D models, dynamic components & templates, audio and beat matching. And it’s all programmable. Design your own assets directly on the canvas, bring them in from Figma or the web, or just let the agent find them, mock them up, or record what it needs automatically. You can even drop in your screen recordings and just let loose. Supercut users: yes, it’ll have first-class support for your recordings, but it’ll work with anything. I’m going to be dropping a lot more examples and tutorials, so follow along. DM me for early access if you’re willing to give feedback. We’ll open it up to everyone very soon.

Neil

23,042 просмотров • 1 месяц назад

Introducing Kaleido💮 from AI at Meta — a universal generative neural rendering engine for photorealistic, unified object and scene view synthesis. Kaleido is built on a simple but powerful design philosophy: 3D perception is a form of visual common sense. Following this idea, we formulate rendering purely as a sequence-to-sequence generation problem, successfully unifying neural rendering with the architecture principles behind modern language and video models. Unlike traditional neural rendering methods, Kaleido learns 3D purely in a data-driven way, without explicit 3D representations or structures. It acquires spatial understanding directly through large-scale video pretraining, then multi-view 3D data finetuning, inspired by how LLMs acquire textual common sense from large corpora before specialising in domains like coding. Through extensive ablations, we progressively modernised the architecture design and training strategies and tackled key scaling challenges in sequence-to-sequence generative rendering, arriving at a design that’s simple, versatile, and scalable. Kaleido significantly outperforms prior generative models in few-view settings, and remarkably is the first zero-shot generative method matches InstantNGP-level rendering quality in multi-view settings. We view Kaleido also as an alternative step towards world modeling that flexibly spans a spectrum of “realities": with many views, it faithfully reconstructs grounded reality; with fewer views, it imagines plausible unseen details. 🔗 Explore more results and paper:
0:12

Introducing Kaleido💮 from AI at Meta — a universal generative neural rendering engine for photorealistic, unified object and scene view synthesis. Kaleido is built on a simple but powerful design philosophy: 3D perception is a form of visual common sense. Following this idea, we formulate rendering purely as a sequence-to-sequence generation problem, successfully unifying neural rendering with the architecture principles behind modern language and video models. Unlike traditional neural rendering methods, Kaleido learns 3D purely in a data-driven way, without explicit 3D representations or structures. It acquires spatial understanding directly through large-scale video pretraining, then multi-view 3D data finetuning, inspired by how LLMs acquire textual common sense from large corpora before specialising in domains like coding. Through extensive ablations, we progressively modernised the architecture design and training strategies and tackled key scaling challenges in sequence-to-sequence generative rendering, arriving at a design that’s simple, versatile, and scalable. Kaleido significantly outperforms prior generative models in few-view settings, and remarkably is the first zero-shot generative method matches InstantNGP-level rendering quality in multi-view settings. We view Kaleido also as an alternative step towards world modeling that flexibly spans a spectrum of “realities": with many views, it faithfully reconstructs grounded reality; with fewer views, it imagines plausible unseen details. 🔗 Explore more results and paper:

Shikun Liu

22,464 просмотров • 11 месяцев назад

1:05:11 I finished 1st 🥇 on my age group and 5th in the male category. I failed my ambitious goal of sub-60 but I was very happy with my race. And I have a nice goal to chase for the next few months. All the staff at the Hyrox venue knew us! I even talked to the manager of Hyrox Turkey/Dubai/Greece. His name his Mustafa and he let my wife inside the athlete zone to shot videos for our YouTube channel. It’s my first Hyrox and the energy is insane. Everybody is fit, sexy and topless. There’s a DJ and MC so it feels like a gym-club in Bali. No surprise this sport is growing so much! This little experiment is engraved in my head. The last 2 weeks have been surreal. Getting paid $112,000 to race, receiving a fancy car at my hotel lobby, getting recognized by a lot of people, my first race… so many things to process! I want to thank my sponsors who made this possible: - by Higgsfield AI 🧩 (10 spots!) - by Kelvin Cobanaj - by Kimi - by OrcaRouter 🐳 - by Peptide Confidential - by John Hu (🍑) I want to thank my wife for always trusting in me and pushing me to try whatever crosses my mind. I want to thank all of you who supported the idea, online and offline. And I want to thank me for believing in myself. I was lost for so long. I felt misunderstood by people thinking I was weird for trying all kind of experiments. In the end I just want to have fun, and I’m incredibly grateful that my hard work has finally compounded and now I can play and get paid for it. Ok I’m a bit emotional now, and I’m getting late for my second race of the day with my wife. This one is just for fun. Woohoo!
0:29

1:05:11 I finished 1st 🥇 on my age group and 5th in the male category. I failed my ambitious goal of sub-60 but I was very happy with my race. And I have a nice goal to chase for the next few months. All the staff at the Hyrox venue knew us! I even talked to the manager of Hyrox Turkey/Dubai/Greece. His name his Mustafa and he let my wife inside the athlete zone to shot videos for our YouTube channel. It’s my first Hyrox and the energy is insane. Everybody is fit, sexy and topless. There’s a DJ and MC so it feels like a gym-club in Bali. No surprise this sport is growing so much! This little experiment is engraved in my head. The last 2 weeks have been surreal. Getting paid $112,000 to race, receiving a fancy car at my hotel lobby, getting recognized by a lot of people, my first race… so many things to process! I want to thank my sponsors who made this possible: - by Higgsfield AI 🧩 (10 spots!) - by Kelvin Cobanaj - by Kimi - by OrcaRouter 🐳 - by Peptide Confidential - by John Hu (🍑) I want to thank my wife for always trusting in me and pushing me to try whatever crosses my mind. I want to thank all of you who supported the idea, online and offline. And I want to thank me for believing in myself. I was lost for so long. I felt misunderstood by people thinking I was weird for trying all kind of experiments. In the end I just want to have fun, and I’m incredibly grateful that my hard work has finally compounded and now I can play and get paid for it. Ok I’m a bit emotional now, and I’m getting late for my second race of the day with my wife. This one is just for fun. Woohoo!

Marc Lou

276,601 просмотров • 12 часов назад

Congratulations to our winner for the Solo Leveling PC jedi_master_snorlax 🎉 Also, please be kind to winners, we only have 1 PC to go around for all of you 🥹🙏 How are winners picked for giveaways??⬇️ Since we’ve got a lot of new people in here, let me explain how winners are picked. We use a site called Rafflly (I’ll link it below if you want to check it out and use it for your own giveaways), We pay $50 a month for a subscription since we have so many entries, and that lets us handle all the volume fairly and smoothly, we had 334,264 enter this time around. I’ve also included a video that shows how we link all the accounts together and run a multi network giveaway to pick one lucky person. It’s the best way to keep things fair! After it picks someone we confirm 1 that they are real and 2 they followed all the rules. And just to clear things up, our giveaways are definitely real. Josh and I run them ourselves and we cover the shipping costs out of pocket. The companies we partner with aren’t paying for the shipping or paying us to run them, we do it because we want to give back to the community. It’s a win win, you get free stuff, we get a little exposure, and the brands do too and nothing gets sent to a landfill 👌 Also, big thank you to everyone who entered the Solo Leveling giveaway! Right now, the Hatsune Miku giveaway is live, you’ve got 6 days left to enter 💙✨ And on the 13th, our 3rd giveaway will go live, the all white Cougar PC 🤍🔥 see you then! 💜 Thank you again for all the love and support, it genuinely makes all these giveaways possible!
0:09

Congratulations to our winner for the Solo Leveling PC jedi_master_snorlax 🎉 Also, please be kind to winners, we only have 1 PC to go around for all of you 🥹🙏 How are winners picked for giveaways??⬇️ Since we’ve got a lot of new people in here, let me explain how winners are picked. We use a site called Rafflly (I’ll link it below if you want to check it out and use it for your own giveaways), We pay $50 a month for a subscription since we have so many entries, and that lets us handle all the volume fairly and smoothly, we had 334,264 enter this time around. I’ve also included a video that shows how we link all the accounts together and run a multi network giveaway to pick one lucky person. It’s the best way to keep things fair! After it picks someone we confirm 1 that they are real and 2 they followed all the rules. And just to clear things up, our giveaways are definitely real. Josh and I run them ourselves and we cover the shipping costs out of pocket. The companies we partner with aren’t paying for the shipping or paying us to run them, we do it because we want to give back to the community. It’s a win win, you get free stuff, we get a little exposure, and the brands do too and nothing gets sent to a landfill 👌 Also, big thank you to everyone who entered the Solo Leveling giveaway! Right now, the Hatsune Miku giveaway is live, you’ve got 6 days left to enter 💙✨ And on the 13th, our 3rd giveaway will go live, the all white Cougar PC 🤍🔥 see you then! 💜 Thank you again for all the love and support, it genuinely makes all these giveaways possible!

Nay.Tesseract

44,102 просмотров • 8 месяцев назад

Today, I graduated from Stanford with my masters in law, with a focus in law, science, and technology. I expended considerable time exploring how to add value to emerging companies by advising them throughout their lifecycles, specifically on raising venture capital and licensing intellectual properties and technologies. I hope to spend the foreseeable future deepening my expertise in these areas. For the Stanford journey, I must start with gratitude. First, I am grateful for my parents, siblings and family who supported me through everything. I am grateful for my former employers who invested heavily in my growth, played a substantial part in Stanford happening, and in whom I found another family. I am especially grateful for my friends (including those I made at Stanford) who have overwhelmed me with love and support that I do not think I deserve. I am grateful for my colleagues, with whom I have shared this journey, and who have blessed me with their knowledge and experiences. I am grateful for all my mentors and professional connections whose benevolence towards me have never run dry. I am grateful for my professors and the Stanford academic faculty who have shared their knowledge with me and have challenged me intellectually, and finally, I am grateful to God, for the invisible hand in how my life has played out over the past year. At the ceremony, my friend Ana Julia – who spoke on behalf of the class – said we already made it before we arrived for this program. That is true, because the program is a gathering of highly exceptional people from different parts of the world, and with different experiences. Having been privileged to spend the last one year with these people, I will go on from here knowing that I mean something. I am something. And I will devote my time to leading with my humanity and advocating for positive change in the world. I charge you to do the same.
0:14

Today, I graduated from Stanford with my masters in law, with a focus in law, science, and technology. I expended considerable time exploring how to add value to emerging companies by advising them throughout their lifecycles, specifically on raising venture capital and licensing intellectual properties and technologies. I hope to spend the foreseeable future deepening my expertise in these areas. For the Stanford journey, I must start with gratitude. First, I am grateful for my parents, siblings and family who supported me through everything. I am grateful for my former employers who invested heavily in my growth, played a substantial part in Stanford happening, and in whom I found another family. I am especially grateful for my friends (including those I made at Stanford) who have overwhelmed me with love and support that I do not think I deserve. I am grateful for my colleagues, with whom I have shared this journey, and who have blessed me with their knowledge and experiences. I am grateful for all my mentors and professional connections whose benevolence towards me have never run dry. I am grateful for my professors and the Stanford academic faculty who have shared their knowledge with me and have challenged me intellectually, and finally, I am grateful to God, for the invisible hand in how my life has played out over the past year. At the ceremony, my friend Ana Julia – who spoke on behalf of the class – said we already made it before we arrived for this program. That is true, because the program is a gathering of highly exceptional people from different parts of the world, and with different experiences. Having been privileged to spend the last one year with these people, I will go on from here knowing that I mean something. I am something. And I will devote my time to leading with my humanity and advocating for positive change in the world. I charge you to do the same.

ABI

183,294 просмотров • 3 лет назад

Today, I am 30 and saying goodbye to my twenties. As I celebrate this milestone, I want to take a moment to reflect on the incredible journey I've experienced during my twenties. The past decade was a period of profound growth, self-discovery, and transformative experiences that shaped me into who I am today. My twenties were filled with many highs and some lows, but through it all, I was resilient and persistent. I faced challenges head-on and learned important lessons along the way that shaped my perspective and enlightened my understanding of life. One of the most significant aspects of my twenties was the journey of taking back my own happiness and self-discovery. I uncovered my passions, strengths, and true purpose. This journey allowed me to embrace my authentic self and my dreams. I learned to respect myself, my time, and my energy more. I realized the importance of setting healthy boundaries, prioritizing self-care, and surrounding myself with positive individuals in my life. By honoring my needs and valuing my well-being, I have created a solid foundation for personal growth and fulfillment. Part of the process involved removing negative energy and people from my life. I recognized those who didn't genuinely want the best for me. Letting go of toxic relationships allowed me to create the space I wanted for positivity, growth, and healthier connections. Another pivotal moment during my twenties was a career change. Recognizing my desire for alignment between my passion and my profession, I bravely pursued a new path. It was not easy—that decision required courage, determination, and betting on myself. I started small before reaching greater heights, and I embraced that challenge. Eventually, it led me to a career that brings me immense fulfillment and joy. My willingness to embrace change and take calculated risks has set me on a path of not only continued success but also my own goals and personal satisfaction. Amidst all of the experiences, the most precious gifts of my twenties were the arrival of my two beautiful boys. Parenthood has brought immeasurable joy, love, and purpose into my life. Through the joys and challenges of raising children, I have grown in compassion, patience, and the ability to nurture and care for others. My boys are a constant reminder of the love and happiness that fill my world every day. As I celebrate my 30th birthday, I want to express gratitude to my wife, my two sons, the incredible family I have, and the friends who have been there for me throughout my journey. They have witnessed everything. Through it all, their love, support, and shared memories have brought immense richness and depth to my life, reminding me of the power of real connection and the importance of cherishing these relationships I hold so close to my heart. As the next decade begins for me, I will embrace the wisdom and experiences I gained during my twenties. I will trust in my abilities and continue to pursue my dreams with passion and resilience. My thirties hold the promise of further personal growth, new adventures, and the opportunity to create lasting memories with my loved ones. Thank you all for everything! With so much love and gratitude, Geomar Milian
0:13

Today, I am 30 and saying goodbye to my twenties. As I celebrate this milestone, I want to take a moment to reflect on the incredible journey I've experienced during my twenties. The past decade was a period of profound growth, self-discovery, and transformative experiences that shaped me into who I am today. My twenties were filled with many highs and some lows, but through it all, I was resilient and persistent. I faced challenges head-on and learned important lessons along the way that shaped my perspective and enlightened my understanding of life. One of the most significant aspects of my twenties was the journey of taking back my own happiness and self-discovery. I uncovered my passions, strengths, and true purpose. This journey allowed me to embrace my authentic self and my dreams. I learned to respect myself, my time, and my energy more. I realized the importance of setting healthy boundaries, prioritizing self-care, and surrounding myself with positive individuals in my life. By honoring my needs and valuing my well-being, I have created a solid foundation for personal growth and fulfillment. Part of the process involved removing negative energy and people from my life. I recognized those who didn't genuinely want the best for me. Letting go of toxic relationships allowed me to create the space I wanted for positivity, growth, and healthier connections. Another pivotal moment during my twenties was a career change. Recognizing my desire for alignment between my passion and my profession, I bravely pursued a new path. It was not easy—that decision required courage, determination, and betting on myself. I started small before reaching greater heights, and I embraced that challenge. Eventually, it led me to a career that brings me immense fulfillment and joy. My willingness to embrace change and take calculated risks has set me on a path of not only continued success but also my own goals and personal satisfaction. Amidst all of the experiences, the most precious gifts of my twenties were the arrival of my two beautiful boys. Parenthood has brought immeasurable joy, love, and purpose into my life. Through the joys and challenges of raising children, I have grown in compassion, patience, and the ability to nurture and care for others. My boys are a constant reminder of the love and happiness that fill my world every day. As I celebrate my 30th birthday, I want to express gratitude to my wife, my two sons, the incredible family I have, and the friends who have been there for me throughout my journey. They have witnessed everything. Through it all, their love, support, and shared memories have brought immense richness and depth to my life, reminding me of the power of real connection and the importance of cherishing these relationships I hold so close to my heart. As the next decade begins for me, I will embrace the wisdom and experiences I gained during my twenties. I will trust in my abilities and continue to pursue my dreams with passion and resilience. My thirties hold the promise of further personal growth, new adventures, and the opportunity to create lasting memories with my loved ones. Thank you all for everything! With so much love and gratitude, Geomar Milian

Geo Milian

15,654 просмотров • 3 лет назад

I'm joining to build and grow ElevenCreative. I've spent most of my career building for creators. ElevenLabs is one of the fastest-growing AI companies, and it's clear they've been principled about how they build for creators every step of the way: → $11M+ paid back to creators through a voice marketplace that lets you control how your voice is used and earn from it → A music product setting the ethical standard with 50/50 royalty splits, trained only on licensed data → Leading voice and audio models that support 70+ languages so creators anywhere in the world can reach a global audience and earn on equal footing → Partnerships with 450+ nonprofits and people with ALS, cancer, and stroke reclaim their voice for them and their loved ones → An all-in-one creative workspace that supports everyone from millions of creators to brands like Disney, Duolingo, and Epic Games → Tooling that unlocks new workflows, like enabling indie authors to publish audiobooks where the economics aren't cost-prohibitive, and so much more Excited to work with @matiii, Luke Harries, and an incredible team that does not stop shipping. If you have feedback on the product, or are excited to join, reach out. Will be hiring for several roles.
0:10

I'm joining to build and grow ElevenCreative. I've spent most of my career building for creators. ElevenLabs is one of the fastest-growing AI companies, and it's clear they've been principled about how they build for creators every step of the way: → $11M+ paid back to creators through a voice marketplace that lets you control how your voice is used and earn from it → A music product setting the ethical standard with 50/50 royalty splits, trained only on licensed data → Leading voice and audio models that support 70+ languages so creators anywhere in the world can reach a global audience and earn on equal footing → Partnerships with 450+ nonprofits and people with ALS, cancer, and stroke reclaim their voice for them and their loved ones → An all-in-one creative workspace that supports everyone from millions of creators to brands like Disney, Duolingo, and Epic Games → Tooling that unlocks new workflows, like enabling indie authors to publish audiobooks where the economics aren't cost-prohibitive, and so much more Excited to work with @matiii, Luke Harries, and an incredible team that does not stop shipping. If you have feedback on the product, or are excited to join, reach out. Will be hiring for several roles.

aneri

60,313 просмотров • 7 месяцев назад