正在加载视频...

视频加载失败

返回首页

Mistral AI Releases Robostral Navigate: An 8B Model Enabling Robots to Navigate Complex Environments Hitting 76.6% on R2R-CE With One RGB Camera. No LiDAR. No depth sensor. No multi-camera rig. Here's how it works. 👇 1. Pointing, not metric commands The model predicts the pixel coordinates of the next...

Marktechpost's profile picture

Marktechpost AI

11,455 subscribers

39,955 次观看 • 1 个月前 •via X (Twitter)

科学技术教育新闻政治
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

While Anthropic and OpenAI race on AI Agents/LLMs, Mistral continues expanding its footprint in physical AI. They just released Robostral Navigate (8B VLA model). It shows strong potential in delivery, logistics, manufacturing, and hospitality--for example, enabling humanoid robots to autonomously navigate multiple rooms to hand a glass of water to an employee,all via natural language instructions in complex environments. Powered by just a single RGB camera (no LiDAR or depth sensors), it achieves 76.6% success on R2R-CE (Room-to-Room in Continuous Environments) unseen benchmarks, outperforming multi-sensor systems. Trained purely in simulation, it generalizes across wheeled, legged, and flying robots.
0:25

While Anthropic and OpenAI race on AI Agents/LLMs, Mistral continues expanding its footprint in physical AI. They just released Robostral Navigate (8B VLA model). It shows strong potential in delivery, logistics, manufacturing, and hospitality--for example, enabling humanoid robots to autonomously navigate multiple rooms to hand a glass of water to an employee,all via natural language instructions in complex environments. Powered by just a single RGB camera (no LiDAR or depth sensors), it achieves 76.6% success on R2R-CE (Room-to-Room in Continuous Environments) unseen benchmarks, outperforming multi-sensor systems. Trained purely in simulation, it generalizes across wheeled, legged, and flying robots.

CyberRobo

10,970 次观看 • 1 个月前

New Course: Post-training of LLMs Learn to post-train and customize an LLM in this short course, taught by Banghua Zhu, Assistant Professor at the University of Washington University of Washington, and co-founder of @NexusflowX. Training an LLM to follow instructions or answer questions has two key stages: pre-training and post-training. In pre-training, it learns to predict the next word or token from large amounts of unlabeled text. In post-training, it learns useful behaviors such as following instructions, tool use, and reasoning. Post-training transforms a general-purpose token predictor—trained on trillions of unlabeled text tokens—into an assistant that follows instructions and performs specific tasks. Because it is much cheaper than pre-training, it is practical for many more teams to incorporate post-training methods into their workflows than pre-training. In this course, you’ll learn three common post-training methods—Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Online Reinforcement Learning (RL)—and how to use each one effectively. With SFT, you train the model on pairs of input and ideal output responses. With DPO, you provide both a preferred (chosen) and a less preferred (rejected) response and train the model to favor the preferred output. With RL, the model generates an output, receives a reward score based on human or automated feedback, and updates the model to improve performance. You’ll learn the basic concepts, common use cases, and principles for curating high-quality data for effective training. Through hands-on labs, you’ll download a pre-trained model from Hugging Face and post-train it using SFT, DPO, and RL to see how each technique shapes model behavior. In detail, you’ll: - Understand what post-training is, when to use it, and how it differs from pre-training. - Build an SFT pipeline to turn a base model into an instruct model. - Explore how DPO reshapes behavior by minimizing contrastive loss—penalizing poor responses and reinforcing preferred ones. - Implement a DPO pipeline to change the identity of a chat assistant. - Learn online RL methods such as Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO), and how to design reward functions. - Train a model with GRPO to improve its math capabilities using a verifiable reward. Post-training is one of the most rapidly developing areas of LLM training. Whether you’re building a high-accuracy context-specific assistant, fine-tuning a model's tone, or improving task-specific accuracy, this course will give you experience with the most important techniques shaping how LLMs are post-trained today. Please sign up here:
2:52

New Course: Post-training of LLMs Learn to post-train and customize an LLM in this short course, taught by Banghua Zhu, Assistant Professor at the University of Washington University of Washington, and co-founder of @NexusflowX. Training an LLM to follow instructions or answer questions has two key stages: pre-training and post-training. In pre-training, it learns to predict the next word or token from large amounts of unlabeled text. In post-training, it learns useful behaviors such as following instructions, tool use, and reasoning. Post-training transforms a general-purpose token predictor—trained on trillions of unlabeled text tokens—into an assistant that follows instructions and performs specific tasks. Because it is much cheaper than pre-training, it is practical for many more teams to incorporate post-training methods into their workflows than pre-training. In this course, you’ll learn three common post-training methods—Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Online Reinforcement Learning (RL)—and how to use each one effectively. With SFT, you train the model on pairs of input and ideal output responses. With DPO, you provide both a preferred (chosen) and a less preferred (rejected) response and train the model to favor the preferred output. With RL, the model generates an output, receives a reward score based on human or automated feedback, and updates the model to improve performance. You’ll learn the basic concepts, common use cases, and principles for curating high-quality data for effective training. Through hands-on labs, you’ll download a pre-trained model from Hugging Face and post-train it using SFT, DPO, and RL to see how each technique shapes model behavior. In detail, you’ll: - Understand what post-training is, when to use it, and how it differs from pre-training. - Build an SFT pipeline to turn a base model into an instruct model. - Explore how DPO reshapes behavior by minimizing contrastive loss—penalizing poor responses and reinforcing preferred ones. - Implement a DPO pipeline to change the identity of a chat assistant. - Learn online RL methods such as Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO), and how to design reward functions. - Train a model with GRPO to improve its math capabilities using a verifiable reward. Post-training is one of the most rapidly developing areas of LLM training. Whether you’re building a high-accuracy context-specific assistant, fine-tuning a model's tone, or improving task-specific accuracy, this course will give you experience with the most important techniques shaping how LLMs are post-trained today. Please sign up here:

Andrew Ng

125,146 次观看 • 1 年前

The robot’s “eyes” just received a big upgrade. LingBot-Depth 2.0, a depth-completion model with half the depth error just dropped. 12/16 benchmarks topped. Glass, mirrors, and transparent objects are so easy for us humans, but so hard for robots, because they do not behave like ordinary surfaces in a camera pipeline. A robot that misunderstands a balcony window or a table edge, will have a completely false planning inside a false world. Huge implecation. LingBot-Depth 2.0 takes an RGB image plus a broken depth map from a sensor and then outputs a cleaner depth map and a usable 3D point cloud. Numbers on LingBot-Depth 2.0 - Excels on glass, mirrors & transparent objects — where traditional depth cameras fail - Training data: 3M → 150M (50x scale-up) - 12 out of 16 first-place rankings on depth completion benchmarks - RMSE cut in half: 0.132 → 0.062 on the hardest indoor scenes LingBot-Vision trained on boundaries, because object edges carry the geometry robots need. No human boundary labels are used, which makes this approach easier to scale. The open-sourced LingBot-Vision is the general vision backbone, and LingBot-Depth 2.0 is the depth model built on it.
2:22

The robot’s “eyes” just received a big upgrade. LingBot-Depth 2.0, a depth-completion model with half the depth error just dropped. 12/16 benchmarks topped. Glass, mirrors, and transparent objects are so easy for us humans, but so hard for robots, because they do not behave like ordinary surfaces in a camera pipeline. A robot that misunderstands a balcony window or a table edge, will have a completely false planning inside a false world. Huge implecation. LingBot-Depth 2.0 takes an RGB image plus a broken depth map from a sensor and then outputs a cleaner depth map and a usable 3D point cloud. Numbers on LingBot-Depth 2.0 - Excels on glass, mirrors & transparent objects — where traditional depth cameras fail - Training data: 3M → 150M (50x scale-up) - 12 out of 16 first-place rankings on depth completion benchmarks - RMSE cut in half: 0.132 → 0.062 on the hardest indoor scenes LingBot-Vision trained on boundaries, because object edges carry the geometry robots need. No human boundary labels are used, which makes this approach easier to scale. The open-sourced LingBot-Vision is the general vision backbone, and LingBot-Depth 2.0 is the depth model built on it.

Rohan Paul

25,312 次观看 • 1 个月前

2D is what the camera sees.👀 3D is what the world is. 🌍 Today, we’re proudly open-sourcing QuerySplat — an open-source feed-forward 3D model that turns a few unposed images into a navigable 3D scene in seconds (run on 4090🪶). Your input: a few unposed images Our output: a sharp, navigable 3D scene — generated in seconds! 🚀 🧠 organize scenes with 3D queries, not pixel-bound Gaussians 📷 no camera poses or manual calibration required ⚡ reconstruct new scenes in a single forward pass ✨ preserve sharp geometry and high-frequency appearance 🏆 achieve state-of-the-art results on DL3DV-Evaluation 🔓 code and model weights are now open-source From pixels to scenes, we are making 3D generation scalable. Try our free iOS app — it’s a lot of fun! 😆 🔗 Try it on the Web: 🔗 Paper: 🔗 Code & weights: 🔗 Project page: 🔗 #AIGC #3DGS #AI #Computervision #3D #artist #3dartist #3d
1:13

2D is what the camera sees.👀 3D is what the world is. 🌍 Today, we’re proudly open-sourcing QuerySplat — an open-source feed-forward 3D model that turns a few unposed images into a navigable 3D scene in seconds (run on 4090🪶). Your input: a few unposed images Our output: a sharp, navigable 3D scene — generated in seconds! 🚀 🧠 organize scenes with 3D queries, not pixel-bound Gaussians 📷 no camera poses or manual calibration required ⚡ reconstruct new scenes in a single forward pass ✨ preserve sharp geometry and high-frequency appearance 🏆 achieve state-of-the-art results on DL3DV-Evaluation 🔓 code and model weights are now open-source From pixels to scenes, we are making 3D generation scalable. Try our free iOS app — it’s a lot of fun! 😆 🔗 Try it on the Web: 🔗 Paper: 🔗 Code & weights: 🔗 Project page: 🔗 #AIGC #3DGS #AI #Computervision #3D #artist #3dartist #3d

InSpatio｜影溯

29,682 次观看 • 29 天前

AI BASKETBALL ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, OPENCV, AND PYTHON. Take any NBA broadcast. Any camera angle. Any resolution. Feed it into the system. YOLO finds every player and the ball. Frame by frame. No manual annotation. No pre-labeled data.The model just sees the court and understands it. Zero-shot classification looks at jersey pixels and decides who plays for whom. Two teams separated in milliseconds. Without ever being told the team names. But here's where it gets wild. Court keypoint detection identifies the geometry of the playing surface. Homography transforms the broadcast camera into a top-down tactical map. Real meters. Real coordinates. From a flat 2D video. Now every player has a position on a real court. Every movement becomes a data point. Every pass becomes a vector. Speed. Distance. Possession time. Pass networks. All extracted from nothing but pixels. No GPS. No chips in the ball. No million-dollar Second Spectrum setup. A Python script and a GPU. The full pipeline is open source. The tutorial walks through every line of code. From detection to transformation to analytics output. Teams used to pay six figures for this data. Now you build it in an afternoon. Computer vision didn't just enter basketball. It democratized it.
5:58:26

AI BASKETBALL ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, OPENCV, AND PYTHON. Take any NBA broadcast. Any camera angle. Any resolution. Feed it into the system. YOLO finds every player and the ball. Frame by frame. No manual annotation. No pre-labeled data.The model just sees the court and understands it. Zero-shot classification looks at jersey pixels and decides who plays for whom. Two teams separated in milliseconds. Without ever being told the team names. But here's where it gets wild. Court keypoint detection identifies the geometry of the playing surface. Homography transforms the broadcast camera into a top-down tactical map. Real meters. Real coordinates. From a flat 2D video. Now every player has a position on a real court. Every movement becomes a data point. Every pass becomes a vector. Speed. Distance. Possession time. Pass networks. All extracted from nothing but pixels. No GPS. No chips in the ball. No million-dollar Second Spectrum setup. A Python script and a GPU. The full pipeline is open source. The tutorial walks through every line of code. From detection to transformation to analytics output. Teams used to pay six figures for this data. Now you build it in an afternoon. Computer vision didn't just enter basketball. It democratized it.

zostaff

230,988 次观看 • 4 个月前

Mistral AI Releases Leanstral 1.5: An Apache-2.0 Lean 4 Code Agent Model Solving 587 of 672 PutnamBench Problems Most AI theorem proving is a language model generating a proof in one shot, with a verifier bolted on at the end to check it. That's autocomplete with a grader — and Mistral just drew a clear line between that and an actual proof agent. They released Leanstral 1.5 — a 119B MoE with 6.5B active parameters, trained as a code agent that lives inside the Lean 4 compiler loop: propose a proof, read the compiler's goals and errors, refine, repeat until it compiles or the budget runs out. Verification isn't the eval here. It's the training signal. Here's what's actually interesting: → Test-time scaling behaves like a dial: PutnamBench Pass@8 climbs 44 → 244 → 493 → 587 solved as the per-attempt token budget moves 50k → 200k → 1M → 4M → 587/672 on PutnamBench at ~$4 per problem, versus an estimated $300+ for Seed-Prover 1.5 high (a 10 H20-days-per-problem budget) → Saturates miniF2F: 100% on both validation and test sets → Two RL environments in training — a multiturn prover, and a raw-filesystem code agent that edits files, runs bash, and queries the Lean language server for live goals and types → Not just math: an Aeneas (Rust → Lean) pipeline flagged 11 genuine bugs across 57 repos, 5 previously unreported — including an integer overflow in datrs/varinteger when (value + 1) hits Std.U64.MAX Apache 2.0 weights, free API endpoint Full analysis: Model weights: Project: Technical Details: Mistral AI Mistral AI for Developers Sophia Yang, Ph.D.
0:25

Mistral AI Releases Leanstral 1.5: An Apache-2.0 Lean 4 Code Agent Model Solving 587 of 672 PutnamBench Problems Most AI theorem proving is a language model generating a proof in one shot, with a verifier bolted on at the end to check it. That's autocomplete with a grader — and Mistral just drew a clear line between that and an actual proof agent. They released Leanstral 1.5 — a 119B MoE with 6.5B active parameters, trained as a code agent that lives inside the Lean 4 compiler loop: propose a proof, read the compiler's goals and errors, refine, repeat until it compiles or the budget runs out. Verification isn't the eval here. It's the training signal. Here's what's actually interesting: → Test-time scaling behaves like a dial: PutnamBench Pass@8 climbs 44 → 244 → 493 → 587 solved as the per-attempt token budget moves 50k → 200k → 1M → 4M → 587/672 on PutnamBench at ~$4 per problem, versus an estimated $300+ for Seed-Prover 1.5 high (a 10 H20-days-per-problem budget) → Saturates miniF2F: 100% on both validation and test sets → Two RL environments in training — a multiturn prover, and a raw-filesystem code agent that edits files, runs bash, and queries the Lean language server for live goals and types → Not just math: an Aeneas (Rust → Lean) pipeline flagged 11 genuine bugs across 57 repos, 5 previously unreported — including an integer overflow in datrs/varinteger when (value + 1) hits Std.U64.MAX Apache 2.0 weights, free API endpoint Full analysis: Model weights: Project: Technical Details: Mistral AI Mistral AI for Developers Sophia Yang, Ph.D.

Marktechpost AI

56,695 次观看 • 2 个月前

Trained a humanoid entirely in a 3D scan of the office. Zero real-world fine-tuning. It just walked in and worked. RL needs hundreds of thousands of attempts, and real robots can't afford to crash. A misjudged gap or a glass door collision breaks hardware and costs hours resetting. So you train in a sim. But sim policies usually train on randomized, untextured geometry; depth is easy to fake. The robot learns structure, not the real world: no materials, no lighting, no idea what anything actually is. RGB cameras carry all of that but training RGB policies in generic fake worlds won’t generalize to the real world. Niantic Spatial 🌎 Scaniverse reconstructs your scan of the real deployment site. One 360° camera walkthrough → photorealistic 3D Gaussian splat at metric scale → collision mesh pulled from the same reconstruction, so vision and physics match exactly. Drops straight into NVIDIA Isaac Sim/Lab, no manual conversion. Flexion simulation-first approach then seamlessly enables the training of RGB-only nav policies inside that reconstruction. With added domain randomization + large image encoders for robustness, this deploys straight to hardware. No real-world fine-tuning. Deployment: months of on-site adaptation → days. Tune into the NVIDIA livestream on 12 August to hear how these companies are closing the sim2real gap: NVIDIA Robotics ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:47

Trained a humanoid entirely in a 3D scan of the office. Zero real-world fine-tuning. It just walked in and worked. RL needs hundreds of thousands of attempts, and real robots can't afford to crash. A misjudged gap or a glass door collision breaks hardware and costs hours resetting. So you train in a sim. But sim policies usually train on randomized, untextured geometry; depth is easy to fake. The robot learns structure, not the real world: no materials, no lighting, no idea what anything actually is. RGB cameras carry all of that but training RGB policies in generic fake worlds won’t generalize to the real world. Niantic Spatial 🌎 Scaniverse reconstructs your scan of the real deployment site. One 360° camera walkthrough → photorealistic 3D Gaussian splat at metric scale → collision mesh pulled from the same reconstruction, so vision and physics match exactly. Drops straight into NVIDIA Isaac Sim/Lab, no manual conversion. Flexion simulation-first approach then seamlessly enables the training of RGB-only nav policies inside that reconstruction. With added domain randomization + large image encoders for robustness, this deploys straight to hardware. No real-world fine-tuning. Deployment: months of on-site adaptation → days. Tune into the NVIDIA livestream on 12 August to hear how these companies are closing the sim2real gap: NVIDIA Robotics ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

119,163 次观看 • 23 天前

For the past 3 months, I've been heads down building and extending our multimodal data capture rig for Physical AI training. What you're seeing here is a single synchronized capture from our system — RGB, depth (stereo IR camera) feed; real-time 3D hand reconstruction, and a 23×20 tactile pressure grid from sensor gloves, all aligned and timestamped together. Getting here took a lot of unglamorous iteration. We reworked the pipeline to deliver clean, gap-free depth at scale after identifying early depth frame drops caused by data throughput bottlenecks in our capture box. We built per-channel calibration models that convert raw resistance signals from our flex sensor gloves into accurate, drift-corrected force values — solving the creep behavior that made a steady grip falsely appear to fade over time. Even though we’ve stayed quiet on socials, word of mouth spread — our rigs have already been deployed in data factories, manufacturing lines, and homes across the US, Europe, China, and Southeast Asia by our customers and ourselves. In the next few days, I'll be sharing more about how we're tackling rights-cleared data capture from the hardware level, and our closed-loop data-model flywheel — using model evaluation results to identify gaps and guide what data our customers need. More soon, stay tuned…
0:26

For the past 3 months, I've been heads down building and extending our multimodal data capture rig for Physical AI training. What you're seeing here is a single synchronized capture from our system — RGB, depth (stereo IR camera) feed; real-time 3D hand reconstruction, and a 23×20 tactile pressure grid from sensor gloves, all aligned and timestamped together. Getting here took a lot of unglamorous iteration. We reworked the pipeline to deliver clean, gap-free depth at scale after identifying early depth frame drops caused by data throughput bottlenecks in our capture box. We built per-channel calibration models that convert raw resistance signals from our flex sensor gloves into accurate, drift-corrected force values — solving the creep behavior that made a steady grip falsely appear to fade over time. Even though we’ve stayed quiet on socials, word of mouth spread — our rigs have already been deployed in data factories, manufacturing lines, and homes across the US, Europe, China, and Southeast Asia by our customers and ourselves. In the next few days, I'll be sharing more about how we're tackling rights-cleared data capture from the hardware level, and our closed-loop data-model flywheel — using model evaluation results to identify gaps and guide what data our customers need. More soon, stay tuned…

Grace Zhang

54,449 次观看 • 2 个月前

The final lecture of my course is an intro to character training! This is a topic that I've been quietly very invested in for ~18 months, as it: * Has potential for high real world impact * Clearly used extensively at frontier labs * Almost no empirical literature exists * More accessible on academic compute This lecture covers what character training is, reviews model specs, constitutions, the differences, the motivations in real world events, some example research papers I like, and open questions in how it relates to post-training/model use generally. Hopefully this brings more people into the field (and reach out if you have questions). It is one of the more research-y chapters in my book, but one that I felt needed the reference. There is still so little, educational content on the topic online. 0:00 Intro 6:22 Part 1: Fundamentals — character, constitutions, and model specs 19:21 Part 2: Character training in practice 23:23 Part 3: Character elicitation without gradient steps 28:03 Part 4: Open questions (and the end of the course) 32:27 The course, complete Thanks for watching. No need to like and subscribe now that the course is done, you definitely wouldn't! h/t to Sharan for leading the technical work I got to do in the space, and Zafir Stojanovski for investing a lot of attention at this book chapter.
32:55

The final lecture of my course is an intro to character training! This is a topic that I've been quietly very invested in for ~18 months, as it: * Has potential for high real world impact * Clearly used extensively at frontier labs * Almost no empirical literature exists * More accessible on academic compute This lecture covers what character training is, reviews model specs, constitutions, the differences, the motivations in real world events, some example research papers I like, and open questions in how it relates to post-training/model use generally. Hopefully this brings more people into the field (and reach out if you have questions). It is one of the more research-y chapters in my book, but one that I felt needed the reference. There is still so little, educational content on the topic online. 0:00 Intro 6:22 Part 1: Fundamentals — character, constitutions, and model specs 19:21 Part 2: Character training in practice 23:23 Part 3: Character elicitation without gradient steps 28:03 Part 4: Open questions (and the end of the course) 32:27 The course, complete Thanks for watching. No need to like and subscribe now that the course is done, you definitely wouldn't! h/t to Sharan for leading the technical work I got to do in the space, and Zafir Stojanovski for investing a lot of attention at this book chapter.

Nathan Lambert

56,783 次观看 • 27 天前

Here's my conversation with Lucas Atkins and the team at Arcee.ai on their path to training and releasing Trinity Large today. From going all in on open models built end to end in the US 6 months ago to having the model in hand is no easy feet. I loved this conversation on how to design a startup around open models and take a bold step to scale it up. I'm openly an Arcee fan, watching them take risk and pull it off. We discuss: - The state (and future) of open vs. closed models, - The business of selling open models for on-prem deployments, - The story of Arcee AI & going “all-in” on this training run, - The ATOM project, - Building frontier model training teams in 6 months, - and other great topics. I really loved this one, and think you well too. Chapters: 00:00:00 Intro: Arcee AI, Trinity Models & Trinity Large 00:08:26 Transitioning a Company to Pre-training 00:13:00 Technical Decisions: Muon and MoE 00:18:41 Scaling and MoE Training Pain 00:23:14 Post-training and RL Strategies 00:28:09 Team Structure and Data Scaling 00:31:31 The Trinity Manifesto: US Open Weights 00:42:31 Specialized Models and Distillation 00:47:12 Infrastructure and Hosting 400B 00:50:53 Open Source as a Business Moat 00:56:31 Predictions: Best Model in 2026 01:02:29 Lightning Round & Conclusions More great open model builder podcasts coming soon!
1:12:15

Here's my conversation with Lucas Atkins and the team at Arcee.ai on their path to training and releasing Trinity Large today. From going all in on open models built end to end in the US 6 months ago to having the model in hand is no easy feet. I loved this conversation on how to design a startup around open models and take a bold step to scale it up. I'm openly an Arcee fan, watching them take risk and pull it off. We discuss: - The state (and future) of open vs. closed models, - The business of selling open models for on-prem deployments, - The story of Arcee AI & going “all-in” on this training run, - The ATOM project, - Building frontier model training teams in 6 months, - and other great topics. I really loved this one, and think you well too. Chapters: 00:00:00 Intro: Arcee AI, Trinity Models & Trinity Large 00:08:26 Transitioning a Company to Pre-training 00:13:00 Technical Decisions: Muon and MoE 00:18:41 Scaling and MoE Training Pain 00:23:14 Post-training and RL Strategies 00:28:09 Team Structure and Data Scaling 00:31:31 The Trinity Manifesto: US Open Weights 00:42:31 Specialized Models and Distillation 00:47:12 Infrastructure and Hosting 400B 00:50:53 Open Source as a Business Moat 00:56:31 Predictions: Best Model in 2026 01:02:29 Lightning Round & Conclusions More great open model builder podcasts coming soon!

Nathan Lambert

26,872 次观看 • 7 个月前

Molmo by Ai2 - Open source SoTA Multimodal (Vision) Language model, beating Claude 3.5 Sonnet, GPT4V and comparable to GPT4o 🔥 They release four model checkpoints: 1. MolmoE-1B, a mixture of experts model with 1B (active) 7B (total) 2. Molmo-7B-O, most open 7B model 3. Molmo-7B-D, demo model 4. Molmo-72B, best model System Architecture > Input: Multi-scale, multi-crop images generated from the original image. > Vision Encoder: OpenAI's ViT-L/14 336px CLIP model, a powerful ViT, encodes images into vision tokens. > Connector: MLP projects tokens to LLM input space, followed by pooling for dimensionality reduction. > LLM: Decoder-only Transformer, various options (OLMo, OLMoE, Qwen2, Mistral, Gemma2, Phi) with diverse scales and openness. Model Variants > Vision Encoder: Consistent ViT-L/14 CLIP model across variants. > LLM: OLMo-7B-1024, OLMoE-1B-7B-0924, Qwen2 (7B, 72B), Mistral 7B, Gemma2 9B, Phi 3 Medium, offering different capacities and openness levels. Training Strategy > Stage 1: Multimodal pre-training for caption generation with new captioning data. > Stage 2: Supervised fine-tuning on a dataset mixture, updating all parameters. > No RLHF involved, Learning rates adjusted based on component types and pre-training status. > All the weights are available on Hugging Face Hub 🤗 > Compatible with Transformers (Remote Code) Kudos Ai2 for such a brilliant and open work! 🐐 Video credits: Allen AI YT Channel
2:15

Molmo by Ai2 - Open source SoTA Multimodal (Vision) Language model, beating Claude 3.5 Sonnet, GPT4V and comparable to GPT4o 🔥 They release four model checkpoints: 1. MolmoE-1B, a mixture of experts model with 1B (active) 7B (total) 2. Molmo-7B-O, most open 7B model 3. Molmo-7B-D, demo model 4. Molmo-72B, best model System Architecture > Input: Multi-scale, multi-crop images generated from the original image. > Vision Encoder: OpenAI's ViT-L/14 336px CLIP model, a powerful ViT, encodes images into vision tokens. > Connector: MLP projects tokens to LLM input space, followed by pooling for dimensionality reduction. > LLM: Decoder-only Transformer, various options (OLMo, OLMoE, Qwen2, Mistral, Gemma2, Phi) with diverse scales and openness. Model Variants > Vision Encoder: Consistent ViT-L/14 CLIP model across variants. > LLM: OLMo-7B-1024, OLMoE-1B-7B-0924, Qwen2 (7B, 72B), Mistral 7B, Gemma2 9B, Phi 3 Medium, offering different capacities and openness levels. Training Strategy > Stage 1: Multimodal pre-training for caption generation with new captioning data. > Stage 2: Supervised fine-tuning on a dataset mixture, updating all parameters. > No RLHF involved, Learning rates adjusted based on component types and pre-training status. > All the weights are available on Hugging Face Hub 🤗 > Compatible with Transformers (Remote Code) Kudos Ai2 for such a brilliant and open work! 🐐 Video credits: Allen AI YT Channel

Vaibhav (VB) Srivastav

80,474 次观看 • 1 年前

Check out our #PAMI paper with code "Dense Continuous-Time Optical Flow from Event Cameras," where we show how to regress *continuous-time* trajectories of every pixel from event cameras alone or events plus frames! The key idea is to iteratively estimate per-pixel polynomials using a recurrent lookup and update scheme. Paper: Code: DOI: We present a method for estimating dense continuous-time optical flow from event data. Traditional dense optical flow methods compute the pixel displacement between two images. Due to missing information, these approaches cannot recover the pixel trajectories in the blind time between two images. We show that it is possible to compute per-pixel, continuous-time optical flow using events from an event camera. Events provide temporally fine-grained information about movement in pixel space due to their asynchronous nature and microsecond response time. We leverage these benefits to predict pixel trajectories densely in continuous time via parameterized Bézier curves. To achieve this, we build a neural network with strong inductive biases for this task: First, we build multiple sequential correlation volumes in time using event data. Second, we use Bézier curves to index these correlation volumes at multiple timestamps along the trajectory. Third, we use the retrieved correlation to update the Bézier curve representations iteratively. Our method can optionally include image pairs to boost performance further. To train and evaluate our model, we introduce a synthetic dataset (MultiFlow) that features moving objects and ground truth trajectories for every pixel. Our quantitative experiments suggest that our method successfully predicts pixel trajectories in continuous time and is competitive in the traditional two-view pixel displacement metric on MultiFlow and DSEC-Flow. Open source code and datasets are released to the public. Kudos to Mathias Gehrig Manasi Muglikar
0:23

Check out our #PAMI paper with code "Dense Continuous-Time Optical Flow from Event Cameras," where we show how to regress *continuous-time* trajectories of every pixel from event cameras alone or events plus frames! The key idea is to iteratively estimate per-pixel polynomials using a recurrent lookup and update scheme. Paper: Code: DOI: We present a method for estimating dense continuous-time optical flow from event data. Traditional dense optical flow methods compute the pixel displacement between two images. Due to missing information, these approaches cannot recover the pixel trajectories in the blind time between two images. We show that it is possible to compute per-pixel, continuous-time optical flow using events from an event camera. Events provide temporally fine-grained information about movement in pixel space due to their asynchronous nature and microsecond response time. We leverage these benefits to predict pixel trajectories densely in continuous time via parameterized Bézier curves. To achieve this, we build a neural network with strong inductive biases for this task: First, we build multiple sequential correlation volumes in time using event data. Second, we use Bézier curves to index these correlation volumes at multiple timestamps along the trajectory. Third, we use the retrieved correlation to update the Bézier curve representations iteratively. Our method can optionally include image pairs to boost performance further. To train and evaluate our model, we introduce a synthetic dataset (MultiFlow) that features moving objects and ground truth trajectories for every pixel. Our quantitative experiments suggest that our method successfully predicts pixel trajectories in continuous time and is competitive in the traditional two-view pixel displacement metric on MultiFlow and DSEC-Flow. Open source code and datasets are released to the public. Kudos to Mathias Gehrig Manasi Muglikar

Davide Scaramuzza

12,681 次观看 • 2 年前

AI TENNIS ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, PYTORCH, AND KEYPOINT EXTRACTION. Take any tennis match broadcast, any camera angle, any resolution. Feed it into the pipeline. YOLO detects both players and the tennis ball frame by frame. No manual labeling, no pre-annotated dataset. A fine-tuned YOLOv5 model trained on a Roboflow tennis ball dataset handles the ball - the hardest object to track in any sport. Tiny, fast, constantly occluded. The model finds it anyway. Trackers maintain identity across frames so Player 1 stays Player 1 from the first serve to match point. But detection is just the start. A ResNet50 CNN trained in PyTorch predicts court keypoints from every frame - the corners, service lines, baselines, net posts. Fourteen points that define the entire playing surface geometry. From those keypoints the system builds a homography matrix and warps the broadcast perspective into a top-down mini court with real coordinates. Now every player has a position in real space, not pixel space. Every frame becomes a measurement. Every rally becomes a dataset. Player movement speed - calculated from position deltas between frames, converted to meters per second through the homography. Ball shot speed - measured from the ball trajectory across consecutive detections. Number of shots per rally - counted automatically through ball direction changes. All of this rendered live on the video as an overlay. A mini court in the corner showing both players as dots moving in real time. Stats updating after every point. OpenCV handles the rendering. Pandas handles the math. PyTorch handles the intelligence. YOLO handles the eyes. No Hawkeye subscription, no court-embedded sensors, no tracking chips in the ball. A Python script, a trained model, and a GPU. The full code is on GitHub. The tutorial walks through every module - from ball detector training to court keypoint extraction to the final statistical overlay. Professional teams used to need broadcast deals and proprietary hardware for this kind of analysis. Now you build it in an afternoon with open-source tools. Trading here: Computer vision didn't just enter tennis. It made the expensive stuff free.
4:41:24

AI TENNIS ANALYSIS. A FULL COMPUTER VISION SYSTEM. BUILT ON YOLO, PYTORCH, AND KEYPOINT EXTRACTION. Take any tennis match broadcast, any camera angle, any resolution. Feed it into the pipeline. YOLO detects both players and the tennis ball frame by frame. No manual labeling, no pre-annotated dataset. A fine-tuned YOLOv5 model trained on a Roboflow tennis ball dataset handles the ball - the hardest object to track in any sport. Tiny, fast, constantly occluded. The model finds it anyway. Trackers maintain identity across frames so Player 1 stays Player 1 from the first serve to match point. But detection is just the start. A ResNet50 CNN trained in PyTorch predicts court keypoints from every frame - the corners, service lines, baselines, net posts. Fourteen points that define the entire playing surface geometry. From those keypoints the system builds a homography matrix and warps the broadcast perspective into a top-down mini court with real coordinates. Now every player has a position in real space, not pixel space. Every frame becomes a measurement. Every rally becomes a dataset. Player movement speed - calculated from position deltas between frames, converted to meters per second through the homography. Ball shot speed - measured from the ball trajectory across consecutive detections. Number of shots per rally - counted automatically through ball direction changes. All of this rendered live on the video as an overlay. A mini court in the corner showing both players as dots moving in real time. Stats updating after every point. OpenCV handles the rendering. Pandas handles the math. PyTorch handles the intelligence. YOLO handles the eyes. No Hawkeye subscription, no court-embedded sensors, no tracking chips in the ball. A Python script, a trained model, and a GPU. The full code is on GitHub. The tutorial walks through every module - from ball detector training to court keypoint extraction to the final statistical overlay. Professional teams used to need broadcast deals and proprietary hardware for this kind of analysis. Now you build it in an afternoon with open-source tools. Trading here: Computer vision didn't just enter tennis. It made the expensive stuff free.

zostaff

120,370 次观看 • 4 个月前

a team of researchers just proved you don't need a bigger model, you need a smarter plan researchers from Tsinghua and South China University of Technology built a framework called Atomic Task Graph. it turned 7B-8B open-source models into GPT-4 competitors on complex agent benchmarks, beating it on two out of three. no fine-tuning. no extra training. zero parameter updates. current AI agents plan in a straight line. step 1, step 2, step 3. when step 4 fails, the whole chain breaks. and the longer the chain gets, the more the model hallucinates because it's reasoning over a ballooning text history. here's how it works. 1. instead of a linear chain, ATG breaks any complex task into a directed graph where subtask inputs and outputs are explicitly mapped 2. it recursively decomposes each subtask until every node is one atomic tool call 3. independent branches run in parallel instead of waiting in line 4. before anything executes, a lightweight "thought experiment" simulates the plan internally to catch bad dependencies and missing steps early 5. when something breaks at runtime, ATG traces the failure to the exact subgraph that caused it and repairs only that piece. validated work stays frozen. the old way meant a failure at step 5 forced a full replan from scratch. hallucinated actions piled up the longer the task ran. ReAct hit a 43% hallucination rate on household tasks. ATG on an 8B Llama model scored 63.65 on ALFWorld. GPT-4 with ReAct scored 41.24 on the same benchmark. hallucinated actions dropped to 12%. those numbers happened because someone stopped throwing compute at the problem and started thinking about how work gets organized. that's the part that gets me. the industry is spending billions on scale. this team spent time on architecture. and the architecture won.
0:23

a team of researchers just proved you don't need a bigger model, you need a smarter plan researchers from Tsinghua and South China University of Technology built a framework called Atomic Task Graph. it turned 7B-8B open-source models into GPT-4 competitors on complex agent benchmarks, beating it on two out of three. no fine-tuning. no extra training. zero parameter updates. current AI agents plan in a straight line. step 1, step 2, step 3. when step 4 fails, the whole chain breaks. and the longer the chain gets, the more the model hallucinates because it's reasoning over a ballooning text history. here's how it works. 1. instead of a linear chain, ATG breaks any complex task into a directed graph where subtask inputs and outputs are explicitly mapped 2. it recursively decomposes each subtask until every node is one atomic tool call 3. independent branches run in parallel instead of waiting in line 4. before anything executes, a lightweight "thought experiment" simulates the plan internally to catch bad dependencies and missing steps early 5. when something breaks at runtime, ATG traces the failure to the exact subgraph that caused it and repairs only that piece. validated work stays frozen. the old way meant a failure at step 5 forced a full replan from scratch. hallucinated actions piled up the longer the task ran. ReAct hit a 43% hallucination rate on household tasks. ATG on an 8B Llama model scored 63.65 on ALFWorld. GPT-4 with ReAct scored 41.24 on the same benchmark. hallucinated actions dropped to 12%. those numbers happened because someone stopped throwing compute at the problem and started thinking about how work gets organized. that's the part that gets me. the industry is spending billions on scale. this team spent time on architecture. and the architecture won.

Alex Veremeyenko

173,351 次观看 • 1 个月前

There is a beautiful story that just happened in AI so let me share it for a lighter tone weekend post among all the doom stories in our AI field this week. It’s a story of people on three continents building and sharing in the open a new small efficient and state-of-the-art AI model. It started a couple of months ago when a new team in the AI scene released their first model from their headquarters in Paris (France): Mistral 7B. Impressive model, small and very strong performances in the benchmarks, better than all previous models of this size. And open source! So you could build on top of it. Lewis in Bern (Switzerland) and Ed (in Lyon, in the South of France) both from the H4 team, a team of researchers in model fine-tuning and alignment were talking about it over a coffee, in one of these gatherings that often happen at Hugging Face to break the distance between people (literal distance as HF is a remote company). What about fine-tuning it using this new DPO method that a research team from Stanford in California just posted on Arxiv, says one? Hey, that’s a great idea, replies the other. We've just build a great code base (with Nathan, Nazneen, Costa, Younes and all the H4 team and TRL community) let's use it! The next day they start diving in the datasets openly shared on the HF hub and stumble upon two interesting large and good quality fine-tuning datasets recently open-sourced by OpenBMB, a Chinese team from Tsinghua: UltraFeedback and UltraChat. A few rounds of training experiments confirm the intuition, the resulting model is super strong, by far the strongest they have ever seen in their benchmarks from Berkeley and Stanford (LMSYS and Alpaca). Join Clementine, the big boss of the open evaluation leaderboard. Her deep dive into the model capabilities confirms the results: impressive performance. But the H4 team also hosts a famous faculty member, Pr. Sasha Rush, Associate Professor at Cornell University in his daytime, hacker at HF in his nighttime. Joining the conversation, he proposes to quickly draft a research paper to organize and share all the details with the community. A few days later, the model, called Zephyr (a wind like Mistral), paper, and all details are shared with the world. Quickly other companies, everywhere in the world starts to use it. LlamaIndex, a famous data framework and community, shares how the model blew their expectations on real-life use-case benchmarks, while researchers and practitioners discuss the paper and work on the Hugging Face hub. All this happened in just a few weeks catalyzed by open access to knowledge, models, research, and datasets released all over the world (Europe, California, China) and by the idea that people can build upon one another work in AI to bring real-world value with efficient and open models. Stories like this are numerous everywhere around us and make me really proud of the AI community and see how we can build amazingly useful things together. [the video is just me reading this Friday post hahah]
4:30

There is a beautiful story that just happened in AI so let me share it for a lighter tone weekend post among all the doom stories in our AI field this week. It’s a story of people on three continents building and sharing in the open a new small efficient and state-of-the-art AI model. It started a couple of months ago when a new team in the AI scene released their first model from their headquarters in Paris (France): Mistral 7B. Impressive model, small and very strong performances in the benchmarks, better than all previous models of this size. And open source! So you could build on top of it. Lewis in Bern (Switzerland) and Ed (in Lyon, in the South of France) both from the H4 team, a team of researchers in model fine-tuning and alignment were talking about it over a coffee, in one of these gatherings that often happen at Hugging Face to break the distance between people (literal distance as HF is a remote company). What about fine-tuning it using this new DPO method that a research team from Stanford in California just posted on Arxiv, says one? Hey, that’s a great idea, replies the other. We've just build a great code base (with Nathan, Nazneen, Costa, Younes and all the H4 team and TRL community) let's use it! The next day they start diving in the datasets openly shared on the HF hub and stumble upon two interesting large and good quality fine-tuning datasets recently open-sourced by OpenBMB, a Chinese team from Tsinghua: UltraFeedback and UltraChat. A few rounds of training experiments confirm the intuition, the resulting model is super strong, by far the strongest they have ever seen in their benchmarks from Berkeley and Stanford (LMSYS and Alpaca). Join Clementine, the big boss of the open evaluation leaderboard. Her deep dive into the model capabilities confirms the results: impressive performance. But the H4 team also hosts a famous faculty member, Pr. Sasha Rush, Associate Professor at Cornell University in his daytime, hacker at HF in his nighttime. Joining the conversation, he proposes to quickly draft a research paper to organize and share all the details with the community. A few days later, the model, called Zephyr (a wind like Mistral), paper, and all details are shared with the world. Quickly other companies, everywhere in the world starts to use it. LlamaIndex, a famous data framework and community, shares how the model blew their expectations on real-life use-case benchmarks, while researchers and practitioners discuss the paper and work on the Hugging Face hub. All this happened in just a few weeks catalyzed by open access to knowledge, models, research, and datasets released all over the world (Europe, California, China) and by the idea that people can build upon one another work in AI to bring real-world value with efficient and open models. Stories like this are numerous everywhere around us and make me really proud of the AI community and see how we can build amazingly useful things together. [the video is just me reading this Friday post hahah]

Thomas Wolf

169,200 次观看 • 2 年前

#GoProMISSION1 PRO Deep Dive—Part 2 of 6 🎥 A new, cutting-edge 1-inch sensor means incredible dynamic range, even in the darkest environments. Low-light performance like you've never seen from a GoPro, with up to 14 stops of dynamic range at the sensor. Think rich shadow detail, crispy highlights, + true-to-life color with minimal noise and no AI artifacts. The new 1-inch sensor gathers 77% more light per pixel than the competition at 8K, holding detail where other cameras give up. For the darkest scenes, Low Light Mode switches the camera to 4K for Quad Bayer capture, combining 4 pixels into 1 to quadruple each pixel's light-gathering area. Add HLG HDR with dual gain readout—capturing multiple exposures simultaneously to eliminate the ghosting you'd normally get from sequential bracketed HDR—and the result is clean, cinematic footage from high noon to pitch black. Then we take it one step further with 3 built-in Night Effects Modes: Star Trails, Vehicle Light Trails, + Light Painting to change the way you see the world after dark. All of this, in a camera over 85% smaller by volume and over 75% less expensive than the next camera in the 8K60 class. Shop MISSION 1 PRO at or at your local retailer 🛒 #GoPro #Cinematic #Videography
0:43

#GoProMISSION1 PRO Deep Dive—Part 2 of 6 🎥 A new, cutting-edge 1-inch sensor means incredible dynamic range, even in the darkest environments. Low-light performance like you've never seen from a GoPro, with up to 14 stops of dynamic range at the sensor. Think rich shadow detail, crispy highlights, + true-to-life color with minimal noise and no AI artifacts. The new 1-inch sensor gathers 77% more light per pixel than the competition at 8K, holding detail where other cameras give up. For the darkest scenes, Low Light Mode switches the camera to 4K for Quad Bayer capture, combining 4 pixels into 1 to quadruple each pixel's light-gathering area. Add HLG HDR with dual gain readout—capturing multiple exposures simultaneously to eliminate the ghosting you'd normally get from sequential bracketed HDR—and the result is clean, cinematic footage from high noon to pitch black. Then we take it one step further with 3 built-in Night Effects Modes: Star Trails, Vehicle Light Trails, + Light Painting to change the way you see the world after dark. All of this, in a camera over 85% smaller by volume and over 75% less expensive than the next camera in the 8K60 class. Shop MISSION 1 PRO at or at your local retailer 🛒 #GoPro #Cinematic #Videography

GoPro

10,781 次观看 • 1 个月前

Today, we give robots a /skills library that self-evolves and compounds indefinitely! Introducing ASPIRE: a robot solving its 100th task is no longer as clueless as solving its first. Coding agents observe multimodal sensory traces from simulation and real robots, launch an evolutionary search over control programs, and distill the best know-how into an ever-expanding library. ASPIRE is a new type of continual learning: "training" is skill refinement instead of gradient descent. "Trained model" is a repo of sensorimotor skills instead of floating weights. “Distributed training” is a panel of agents each practicing a different skill instead of sharded minibatches. Here's the beauty: ASPIRE gives the tired terms "sim2real transfer" and "cross-embodiment transfer" a whole new meaning. Bridging the sim-to-real gap is notoriously brutal. An end-to-end policy has to swallow both the visual shift (sim looks toyish next to a real camera) and the subtle contact physics it never quite gets right. ASPIRE sidesteps the mess, because it doesn't ship pixels or weights across the gap, but ships the know-how. The robot still has to practice in the real world, not zero-shot, but it gets there way faster because it isn't rediscovering the strategy from scratch. Same for going single-arm to bimanual hardware, which usually requires new data and retraining from zero. ASPIRE achieves up to ~10x cut in "transfer learning” tokens (yes, tokens are the new unit of *training* compute ;) Check out our gallery of 150+ tasks and 90+ skills the robots taught themselves, all on the website! Kind of wild that we can ship the "learned weights" as an HTML page rather than a GGUF. We'll open-source the full stack so your own robot library starts compounding from ours! Deep dive in thread:
0:40

Today, we give robots a /skills library that self-evolves and compounds indefinitely! Introducing ASPIRE: a robot solving its 100th task is no longer as clueless as solving its first. Coding agents observe multimodal sensory traces from simulation and real robots, launch an evolutionary search over control programs, and distill the best know-how into an ever-expanding library. ASPIRE is a new type of continual learning: "training" is skill refinement instead of gradient descent. "Trained model" is a repo of sensorimotor skills instead of floating weights. “Distributed training” is a panel of agents each practicing a different skill instead of sharded minibatches. Here's the beauty: ASPIRE gives the tired terms "sim2real transfer" and "cross-embodiment transfer" a whole new meaning. Bridging the sim-to-real gap is notoriously brutal. An end-to-end policy has to swallow both the visual shift (sim looks toyish next to a real camera) and the subtle contact physics it never quite gets right. ASPIRE sidesteps the mess, because it doesn't ship pixels or weights across the gap, but ships the know-how. The robot still has to practice in the real world, not zero-shot, but it gets there way faster because it isn't rediscovering the strategy from scratch. Same for going single-arm to bimanual hardware, which usually requires new data and retraining from zero. ASPIRE achieves up to ~10x cut in "transfer learning” tokens (yes, tokens are the new unit of *training* compute ;) Check out our gallery of 150+ tasks and 90+ skills the robots taught themselves, all on the website! Kind of wild that we can ship the "learned weights" as an HTML page rather than a GGUF. We'll open-source the full stack so your own robot library starts compounding from ours! Deep dive in thread:

Jim Fan

209,534 次观看 • 2 个月前

i rebuilt the Odyssey the way Homer actually wrote it and handed the whole thing to AI: the gods, the monsters, the fights the movies always skip, scored like the real thing... but it's not, the whole thing was one-shotted by AI in an hour, here's how you can do the same: > access Higgsfield Supercomputer: all image & video models, plus all the skills you need > feed 5-10 stills from a film you love to a vision model and lock its answer into a style contract > build it as 14-16 isolated shots of 3-5 seconds, never one continuous scene, identity drift starts past 30 seconds > animate stills, not ideas: every clip starts from a locked keyframe with a named camera move and an in-scene event > chain shots: the last frame of one clip becomes the reference of the next, continuity for free > time-code the climax inside one generation: [0-4s] wide, [4-8s] push in, [8-12s] close up, [12-15s] reveal > brief the score in movements (build, peak, quiet, resolve), then trim the cut to the music the full system, prompts included, is in the article:
1:29

i rebuilt the Odyssey the way Homer actually wrote it and handed the whole thing to AI: the gods, the monsters, the fights the movies always skip, scored like the real thing... but it's not, the whole thing was one-shotted by AI in an hour, here's how you can do the same: > access Higgsfield Supercomputer: all image & video models, plus all the skills you need > feed 5-10 stills from a film you love to a vision model and lock its answer into a style contract > build it as 14-16 isolated shots of 3-5 seconds, never one continuous scene, identity drift starts past 30 seconds > animate stills, not ideas: every clip starts from a locked keyframe with a named camera move and an in-scene event > chain shots: the last frame of one clip becomes the reference of the next, continuity for free > time-code the climax inside one generation: [0-4s] wide, [4-8s] push in, [8-12s] close up, [12-15s] reveal > brief the score in movements (build, peak, quiet, resolve), then trim the cut to the music the full system, prompts included, is in the article:

Machina

50,694 次观看 • 1 个月前

WATCH ME BUILD AN AI BUSINESS FROM SCRATCH (EPISODE 1) I'm building an AI services business from scratch, on camera, with 4 rules: -100% cold outreach -No audience -No network -60 minutes a day max The scoreboard is how many days it takes a total stranger to pay me $500. Here's the entire model: 1) The niche is tree services in Charlotte. High ticket, no private equity, and after-hours emergencies where the first person to answer wins the job. 2) The offer: a built and hosted website, an AI-powered speed-to-quote form, and a Google Business Profile cleanup. $500 flat. 3) The website IS the lead magnet. We build it first, give it away free, and charge $500 to install and host it. 4) Five agents inside Hyperagent run everything. Prospector, Opener, Sitesmith, Fulfiller, and an Account Manager upsell. 5) The Prospector dispatched 8 subagents, scraped 494 businesses, and verified 98 real leads. Done in 30 minutes. 6) The competition is a joke. The first five "websites" on my list: a blank page, a 404, two dead links, and a redirect to a crypto gaming site. 7) First-touch texts: under 160 characters, no links, no emojis, no mention of AI. "Hey, do y'all still do tree removal in Charlotte? Found you on Google. Corey." 8) The agent wrote "y'all" on its own. Sounding like a neighbor, not a company, is the whole reply-rate game. 9) 60 minutes a day covers 25-50 outreach attempts. Outreach is the only work that actually makes money. 10) All 5 agents get given away free in episode 4, plus $1,000 in free Hyperagent credits at the link in the YouTube subscription. Enough to run my entire lead scrape 6 times over. Two things that I'm looking to accomplish here: 1) Strip away every advantage and prove the model still works. If it works with no audience and one hour a day, nobody has an excuse left. 2) Give away the finished deliverable, not a pitch. An owner looking at their own new website doesn't need convincing. Full breakdown below. (also available on the Build With AI podcast)
43:54

WATCH ME BUILD AN AI BUSINESS FROM SCRATCH (EPISODE 1) I'm building an AI services business from scratch, on camera, with 4 rules: -100% cold outreach -No audience -No network -60 minutes a day max The scoreboard is how many days it takes a total stranger to pay me $500. Here's the entire model: 1) The niche is tree services in Charlotte. High ticket, no private equity, and after-hours emergencies where the first person to answer wins the job. 2) The offer: a built and hosted website, an AI-powered speed-to-quote form, and a Google Business Profile cleanup. $500 flat. 3) The website IS the lead magnet. We build it first, give it away free, and charge $500 to install and host it. 4) Five agents inside Hyperagent run everything. Prospector, Opener, Sitesmith, Fulfiller, and an Account Manager upsell. 5) The Prospector dispatched 8 subagents, scraped 494 businesses, and verified 98 real leads. Done in 30 minutes. 6) The competition is a joke. The first five "websites" on my list: a blank page, a 404, two dead links, and a redirect to a crypto gaming site. 7) First-touch texts: under 160 characters, no links, no emojis, no mention of AI. "Hey, do y'all still do tree removal in Charlotte? Found you on Google. Corey." 8) The agent wrote "y'all" on its own. Sounding like a neighbor, not a company, is the whole reply-rate game. 9) 60 minutes a day covers 25-50 outreach attempts. Outreach is the only work that actually makes money. 10) All 5 agents get given away free in episode 4, plus $1,000 in free Hyperagent credits at the link in the YouTube subscription. Enough to run my entire lead scrape 6 times over. Two things that I'm looking to accomplish here: 1) Strip away every advantage and prove the model still works. If it works with no audience and one hour a day, nobody has an excuse left. 2) Give away the finished deliverable, not a pitch. An owner looking at their own new website doesn't need convincing. Full breakdown below. (also available on the Build With AI podcast)

Corey Ganim

39,767 次观看 • 1 个月前

100 years. 10 iconic looks. One seamless journey through the evolution of men's fashion. Which decade would you wear? 👔✨ Created with GPT image 2.0 and Seedance 2.0 on Thank You AI PROMPT: Style: Ultra-realistic cinematic fashion reel, luxury editorial, premium Instagram creator aesthetic, smooth camera movement, natural lighting, 4K, highly detailed fabrics, realistic clothing physics. Scene Overview A stylish male model showcases five iconic fashion eras from the 1920s to the 1960s. The video takes place in the same modern studio with soft daylight and a clean architectural background. The same model appears throughout the video while only the clothing changes between decades. Every outfit change happens naturally during camera movement, creating a smooth and satisfying visual flow. The only on-screen text is the decade name. Music is a modern electronic fashion track that builds continuously without vocals. Character Use the reference image as the model. Young adult male. Confident but relaxed. Friendly expression. Natural walking pace. The same model appears in every scene. Environment Minimalist fashion studio. Large windows. Warm daylight. Neutral beige walls. Polished concrete floor. Luxury editorial atmosphere. Keep the background consistent throughout the video. 0:00–0:03 — 1920s The video opens with the model standing confidently in the center of the studio. The camera slowly moves closer. He adjusts his cufflinks before looking toward the camera. Outfit Charcoal pinstripe three-piece suit White dress shirt Dark tie Waistcoat Pocket watch chain Oxford shoes Grey newsboy cap On-screen text: 1920s As the camera quickly pans to the right, the clothing smoothly changes into the next decade. 0:03–0:06 — 1930s The camera finishes the movement. The model now wears a cream double-breasted suit with wide lapels, pleated trousers, spectator shoes, and a fedora. He places one hand in his pocket and takes a confident step forward. On-screen text: 1930s The camera gently circles around him, and during the movement the outfit changes into the next style. 0:06–0:09 — 1940s The camera completes the circle. The model now wears an olive green utility jacket with tailored trousers, brown leather boots, and a classic wristwatch. He buttons the jacket while walking slowly toward the camera. On-screen text: 1940s A warm light sweep crosses the frame, leading naturally into the next outfit. 0:09–0:12 — 1950s The model now wears a fitted white T-shirt, dark blue cuffed jeans, a brown leather belt, leather boots, and classic sunglasses. He casually adjusts the sunglasses before taking two relaxed steps forward. On-screen text: 1950s The camera briefly moves closer, then pulls back, allowing the clothing to change smoothly. 0:12–0:15 — 1960s The model now wears a black turtleneck, camel overcoat, slim trousers, Chelsea boots, and a simple wristwatch. He walks confidently toward the camera with both hands in his coat pockets. On-screen text: 1960s During the final second, the camera begins a slow clockwise circle around him. His coat moves naturally as he lightly adjusts the collar. The music continues to build instead of ending. 0:15–0:18 — 1970s The video begins exactly where Part 1 ended. The camera continues its slow clockwise movement around the model before settling into a smooth front-facing angle. The music hits the next beat naturally. Outfit Brown suede jacket Patterned open-collar shirt Flared trousers Brown leather belt Platform shoes Gold wristwatch Aviator sunglasses The model adjusts his sunglasses before removing them with a confident smile. On-screen text: 1970s As the camera moves slightly closer, the outfit changes naturally into the next decade. 0:18–0:21 — 1980s The camera continues tracking forward. The model now wears a charcoal oversized business suit with a patterned tie, pleated trousers, polished loafers, and a classic dress watch. He straightens his tie while walking confidently toward the camera. On-screen text: 1980s The camera briefly passes beside him, creating a smooth transition into the next outfit. 0:21–0:24 — 1990s The camera emerges on the opposite side. The model now wears an oversized light-wash denim jacket over a graphic T-shirt, relaxed jeans, chunky white sneakers, and a simple silver chain. He smiles naturally while walking across the frame with both hands in his jacket pockets. On-screen text: 1990s The camera follows his movement before quickly swinging to reveal the next decade. 0:24–0:27 — 2000s The camera settles into a smooth tracking shot. The model now wears a black leather jacket over a fitted white T-shirt, slim dark jeans, white high-top sneakers, and aviator sunglasses. He casually removes the jacket from one shoulder while continuing to walk. The movement of the jacket naturally fills part of the frame as the clothing changes into the final look. On-screen text: 2000s 0:27–0:30 — 2020s The jacket clears the frame. The camera slowly pushes toward the model. He now wears a modern quiet luxury outfit. Outfit Beige overshirt Premium white T-shirt Relaxed tailored trousers White minimalist sneakers Luxury stainless steel watch Simple silver ring The model walks toward the camera before stopping naturally. He smiles confidently while looking directly into the lens. On-screen text: 2020s The decade text gently fades into: 100 YEARS OF MEN'S FASHION Then below it: 1920 → 2020 The music reaches its final peak before fading smoothly. The camera holds on the model for a brief moment before fading to black. Camera Style Smooth cinematic movement Gentle dolly shots Slow orbit around the model Occasional push-in Natural handheld feel Soft depth of field Luxury fashion commercial look Visual Style Premium editorial photography Warm natural colors Soft contrast Highly detailed fabrics Clean tailoring Realistic lighting Crisp facial details Natural movement Elegant pacing
0:30

100 years. 10 iconic looks. One seamless journey through the evolution of men's fashion. Which decade would you wear? 👔✨ Created with GPT image 2.0 and Seedance 2.0 on Thank You AI PROMPT: Style: Ultra-realistic cinematic fashion reel, luxury editorial, premium Instagram creator aesthetic, smooth camera movement, natural lighting, 4K, highly detailed fabrics, realistic clothing physics. Scene Overview A stylish male model showcases five iconic fashion eras from the 1920s to the 1960s. The video takes place in the same modern studio with soft daylight and a clean architectural background. The same model appears throughout the video while only the clothing changes between decades. Every outfit change happens naturally during camera movement, creating a smooth and satisfying visual flow. The only on-screen text is the decade name. Music is a modern electronic fashion track that builds continuously without vocals. Character Use the reference image as the model. Young adult male. Confident but relaxed. Friendly expression. Natural walking pace. The same model appears in every scene. Environment Minimalist fashion studio. Large windows. Warm daylight. Neutral beige walls. Polished concrete floor. Luxury editorial atmosphere. Keep the background consistent throughout the video. 0:00–0:03 — 1920s The video opens with the model standing confidently in the center of the studio. The camera slowly moves closer. He adjusts his cufflinks before looking toward the camera. Outfit Charcoal pinstripe three-piece suit White dress shirt Dark tie Waistcoat Pocket watch chain Oxford shoes Grey newsboy cap On-screen text: 1920s As the camera quickly pans to the right, the clothing smoothly changes into the next decade. 0:03–0:06 — 1930s The camera finishes the movement. The model now wears a cream double-breasted suit with wide lapels, pleated trousers, spectator shoes, and a fedora. He places one hand in his pocket and takes a confident step forward. On-screen text: 1930s The camera gently circles around him, and during the movement the outfit changes into the next style. 0:06–0:09 — 1940s The camera completes the circle. The model now wears an olive green utility jacket with tailored trousers, brown leather boots, and a classic wristwatch. He buttons the jacket while walking slowly toward the camera. On-screen text: 1940s A warm light sweep crosses the frame, leading naturally into the next outfit. 0:09–0:12 — 1950s The model now wears a fitted white T-shirt, dark blue cuffed jeans, a brown leather belt, leather boots, and classic sunglasses. He casually adjusts the sunglasses before taking two relaxed steps forward. On-screen text: 1950s The camera briefly moves closer, then pulls back, allowing the clothing to change smoothly. 0:12–0:15 — 1960s The model now wears a black turtleneck, camel overcoat, slim trousers, Chelsea boots, and a simple wristwatch. He walks confidently toward the camera with both hands in his coat pockets. On-screen text: 1960s During the final second, the camera begins a slow clockwise circle around him. His coat moves naturally as he lightly adjusts the collar. The music continues to build instead of ending. 0:15–0:18 — 1970s The video begins exactly where Part 1 ended. The camera continues its slow clockwise movement around the model before settling into a smooth front-facing angle. The music hits the next beat naturally. Outfit Brown suede jacket Patterned open-collar shirt Flared trousers Brown leather belt Platform shoes Gold wristwatch Aviator sunglasses The model adjusts his sunglasses before removing them with a confident smile. On-screen text: 1970s As the camera moves slightly closer, the outfit changes naturally into the next decade. 0:18–0:21 — 1980s The camera continues tracking forward. The model now wears a charcoal oversized business suit with a patterned tie, pleated trousers, polished loafers, and a classic dress watch. He straightens his tie while walking confidently toward the camera. On-screen text: 1980s The camera briefly passes beside him, creating a smooth transition into the next outfit. 0:21–0:24 — 1990s The camera emerges on the opposite side. The model now wears an oversized light-wash denim jacket over a graphic T-shirt, relaxed jeans, chunky white sneakers, and a simple silver chain. He smiles naturally while walking across the frame with both hands in his jacket pockets. On-screen text: 1990s The camera follows his movement before quickly swinging to reveal the next decade. 0:24–0:27 — 2000s The camera settles into a smooth tracking shot. The model now wears a black leather jacket over a fitted white T-shirt, slim dark jeans, white high-top sneakers, and aviator sunglasses. He casually removes the jacket from one shoulder while continuing to walk. The movement of the jacket naturally fills part of the frame as the clothing changes into the final look. On-screen text: 2000s 0:27–0:30 — 2020s The jacket clears the frame. The camera slowly pushes toward the model. He now wears a modern quiet luxury outfit. Outfit Beige overshirt Premium white T-shirt Relaxed tailored trousers White minimalist sneakers Luxury stainless steel watch Simple silver ring The model walks toward the camera before stopping naturally. He smiles confidently while looking directly into the lens. On-screen text: 2020s The decade text gently fades into: 100 YEARS OF MEN'S FASHION Then below it: 1920 → 2020 The music reaches its final peak before fading smoothly. The camera holds on the model for a brief moment before fading to black. Camera Style Smooth cinematic movement Gentle dolly shots Slow orbit around the model Occasional push-in Natural handheld feel Soft depth of field Luxury fashion commercial look Visual Style Premium editorial photography Warm natural colors Soft contrast Highly detailed fabrics Clean tailoring Realistic lighting Crisp facial details Natural movement Elegant pacing

Caden Flux

30,445 次观看 • 1 个月前